Make vc-toggle-read-only an alias for toggle-read-only.
[emacs.git] / lisp / international / mule.el
blobcac37b3b8c5b7f3007a9fe6610297d678a426302
1 ;;; mule.el --- basic commands for multilingual environment
3 ;; Copyright (C) 1997-2011 Free Software Foundation, Inc.
4 ;; Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
5 ;; 2005, 2006, 2007, 2008, 2009, 2010, 2011
6 ;; National Institute of Advanced Industrial Science and Technology (AIST)
7 ;; Registration Number H14PRO021
8 ;; Copyright (C) 2003
9 ;; National Institute of Advanced Industrial Science and Technology (AIST)
10 ;; Registration Number H13PRO009
12 ;; Keywords: mule, multilingual, character set, coding system
14 ;; This file is part of GNU Emacs.
16 ;; GNU Emacs is free software: you can redistribute it and/or modify
17 ;; it under the terms of the GNU General Public License as published by
18 ;; the Free Software Foundation, either version 3 of the License, or
19 ;; (at your option) any later version.
21 ;; GNU Emacs is distributed in the hope that it will be useful,
22 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
23 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
24 ;; GNU General Public License for more details.
26 ;; You should have received a copy of the GNU General Public License
27 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
29 ;;; Commentary:
31 ;;; Code:
33 (defconst mule-version "6.0 (HANACHIRUSATO)" "\
34 Version number and name of this version of MULE (multilingual environment).")
36 (defconst mule-version-date "2003.9.1" "\
37 Distribution date of this version of MULE (multilingual environment).")
40 ;;; CHARSET
42 ;; Backward compatibility code for handling emacs-mule charsets.
43 (defvar private-char-area-1-min #xF0000)
44 (defvar private-char-area-1-max #xFFFFE)
45 (defvar private-char-area-2-min #x100000)
46 (defvar private-char-area-2-max #x10FFFE)
48 ;; Table of emacs-mule charsets indexed by their emacs-mule ID.
49 (defvar emacs-mule-charset-table (make-vector 256 nil))
50 (aset emacs-mule-charset-table 0 'ascii)
52 ;; Convert the argument of old-style call of define-charset to a
53 ;; property list used by the new-style.
54 ;; INFO-VECTOR is a vector of the format:
55 ;; [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE
56 ;; SHORT-NAME LONG-NAME DESCRIPTION]
58 (defun convert-define-charset-argument (emacs-mule-id info-vector)
59 (let* ((dim (aref info-vector 0))
60 (chars (aref info-vector 1))
61 (total (if (= dim 1) chars (* chars chars)))
62 (code-space (if (= dim 1) (if (= chars 96) [32 127] [33 126])
63 (if (= chars 96) [32 127 32 127] [33 126 33 126])))
64 code-offset)
65 (if (integerp emacs-mule-id)
66 (or (= emacs-mule-id 0)
67 (and (>= emacs-mule-id 129) (< emacs-mule-id 256))
68 (error "Invalid CHARSET-ID: %d" emacs-mule-id))
69 (let (from-id to-id)
70 (if (= dim 1) (setq from-id 160 to-id 224)
71 (setq from-id 224 to-id 255))
72 (while (and (< from-id to-id)
73 (not (aref emacs-mule-charset-table from-id)))
74 (setq from-id (1+ from-id)))
75 (if (= from-id to-id)
76 (error "No more room for the new Emacs-mule charset"))
77 (setq emacs-mule-id from-id)))
78 (if (> (- private-char-area-1-max private-char-area-1-min) total)
79 (setq code-offset private-char-area-1-min
80 private-char-area-1-min (+ private-char-area-1-min total))
81 (if (> (- private-char-area-2-max private-char-area-2-min) total)
82 (setq code-offset private-char-area-2-min
83 private-char-area-2-min (+ private-char-area-2-min total))
84 (error "No more space for a new charset")))
85 (list :dimension dim
86 :code-space code-space
87 :iso-final-char (aref info-vector 4)
88 :code-offset code-offset
89 :emacs-mule-id emacs-mule-id)))
91 (defun define-charset (name docstring &rest props)
92 "Define NAME (symbol) as a charset with DOCSTRING.
93 The remaining arguments must come in pairs ATTRIBUTE VALUE. ATTRIBUTE
94 may be any symbol. The following have special meanings, and one of
95 `:code-offset', `:map', `:subset', `:superset' must be specified.
97 `:short-name'
99 VALUE must be a short string to identify the charset. If omitted,
100 NAME is used.
102 `:long-name'
104 VALUE must be a string longer than `:short-name' to identify the
105 charset. If omitted, the value of the `:short-name' attribute is used.
107 `:dimension'
109 VALUE must be an integer 0, 1, 2, or 3, specifying the dimension of
110 code-points of the charsets. If omitted, it is calculated from the
111 value of the `:code-space' attribute.
113 `:code-space'
115 VALUE must be a vector of length at most 8 specifying the byte code
116 range of each dimension in this format:
117 [ MIN-1 MAX-1 MIN-2 MAX-2 ... ]
118 where MIN-N is the minimum byte value of Nth dimension of code-point,
119 MAX-N is the maximum byte value of that.
121 `:min-code'
123 VALUE must be an integer specifying the mininum code point of the
124 charset. If omitted, it is calculated from `:code-space'. VALUE may
125 be a cons (HIGH . LOW), where HIGH is the most significant 16 bits of
126 the code point and LOW is the least significant 16 bits.
128 `:max-code'
130 VALUE must be an integer specifying the maxinum code point of the
131 charset. If omitted, it is calculated from `:code-space'. VALUE may
132 be a cons (HIGH . LOW), where HIGH is the most significant 16 bits of
133 the code point and LOW is the least significant 16 bits.
135 `:iso-final-char'
137 VALUE must be a character in the range 32 to 127 (inclusive)
138 specifying the final char of the charset for ISO-2022 encoding. If
139 omitted, the charset can't be encoded by ISO-2022 based
140 coding-systems.
142 `:iso-revision-number'
144 VALUE must be an integer in the range 0..63, specifying the revision
145 number of the charset for ISO-2022 encoding.
147 `:emacs-mule-id'
149 VALUE must be an integer of 0, 129..255. If omitted, the charset
150 can't be encoded by coding-systems of type `emacs-mule'.
152 `:ascii-compatible-p'
154 VALUE must be nil or t (default nil). If VALUE is t, the charset is
155 compatible with ASCII, i.e. the first 128 code points map to ASCII.
157 `:supplementary-p'
159 VALUE must be nil or t. If the VALUE is t, the charset is
160 supplementary, which means it is used only as a parent or a
161 subset of some other charset, or it is provided just for backward
162 compatibility.
164 `:invalid-code'
166 VALUE must be a nonnegative integer that can be used as an invalid
167 code point of the charset. If the minimum code is 0 and the maximum
168 code is greater than Emacs' maximum integer value, `:invalid-code'
169 should not be omitted.
171 `:code-offset'
173 VALUE must be an integer added to the index number of a character to
174 get the corresponding character code.
176 `:map'
178 VALUE must be vector or string.
180 If it is a vector, the format is [ CODE-1 CHAR-1 CODE-2 CHAR-2 ... ],
181 where CODE-n is a code-point of the charset, and CHAR-n is the
182 corresponding character code.
184 If it is a string, it is a name of file that contains the above
185 information. Each line of the file must be this format:
186 0xXXX 0xYYY
187 where XXX is a hexadecimal representation of CODE-n and YYY is a
188 hexadecimal representation of CHAR-n. A line starting with `#' is a
189 comment line.
191 `:subset'
193 VALUE must be a list:
194 ( PARENT MIN-CODE MAX-CODE OFFSET )
195 PARENT is a parent charset. MIN-CODE and MAX-CODE specify the range
196 of characters inherited from the parent. OFFSET is an integer value
197 to add to a code point of the parent charset to get the corresponding
198 code point of this charset.
200 `:superset'
202 VALUE must be a list of parent charsets. The charset inherits
203 characters from them. Each element of the list may be a cons (PARENT
204 . OFFSET), where PARENT is a parent charset, and OFFSET is an offset
205 value to add to a code point of PARENT to get the corresponding code
206 point of this charset.
208 `:unify-map'
210 VALUE must be vector or string.
212 If it is a vector, the format is [ CODE-1 CHAR-1 CODE-2 CHAR-2 ... ],
213 where CODE-n is a code-point of the charset, and CHAR-n is the
214 corresponding Unicode character code.
216 If it is a string, it is a name of file that contains the above
217 information. The file format is the same as what described for `:map'
218 attribute."
219 (when (vectorp (car props))
220 ;; Old style code:
221 ;; (define-charset CHARSET-ID CHARSET-SYMBOL INFO-VECTOR)
222 ;; Convert the argument to make it fit with the current style.
223 (let ((vec (car props)))
224 (setq props (convert-define-charset-argument name vec)
225 name docstring
226 docstring (aref vec 8))))
227 (let ((attrs (mapcar 'list '(:dimension
228 :code-space
229 :min-code
230 :max-code
231 :iso-final-char
232 :iso-revision-number
233 :emacs-mule-id
234 :ascii-compatible-p
235 :supplementary-p
236 :invalid-code
237 :code-offset
238 :map
239 :subset
240 :superset
241 :unify-map
242 :plist))))
244 ;; If :dimension is omitted, get the dimension from :code-space.
245 (let ((dimension (plist-get props :dimension)))
246 (or dimension
247 (let ((code-space (plist-get props :code-space)))
248 (setq dimension (if code-space (/ (length code-space) 2) 4))
249 (setq props (plist-put props :dimension dimension)))))
251 (let ((code-space (plist-get props :code-space)))
252 (or code-space
253 (let ((dimension (plist-get props :dimension)))
254 (setq code-space (make-vector 8 0))
255 (dotimes (i dimension)
256 (aset code-space (1+ (* i 2)) #xFF))
257 (setq props (plist-put props :code-space code-space)))))
259 ;; If :emacs-mule-id is specified, update emacs-mule-charset-table.
260 (let ((emacs-mule-id (plist-get props :emacs-mule-id)))
261 (if (integerp emacs-mule-id)
262 (aset emacs-mule-charset-table emacs-mule-id name)))
264 (dolist (slot attrs)
265 (setcdr slot (purecopy (plist-get props (car slot)))))
267 ;; Make sure that the value of :code-space is a vector of 8
268 ;; elements.
269 (let* ((slot (assq :code-space attrs))
270 (val (cdr slot))
271 (len (length val)))
272 (if (< len 8)
273 (setcdr slot
274 (vconcat val (make-vector (- 8 len) 0)))))
276 ;; Add :name and :docstring properties to PROPS.
277 (setq props
278 (cons :name (cons name (cons :docstring (cons (purecopy docstring) props)))))
279 (or (plist-get props :short-name)
280 (plist-put props :short-name (symbol-name name)))
281 (or (plist-get props :long-name)
282 (plist-put props :long-name (plist-get props :short-name)))
283 (plist-put props :base name)
284 ;; We can probably get a worthwhile amount in purespace.
285 (setq props
286 (mapcar (lambda (elt)
287 (if (stringp elt)
288 (purecopy elt)
289 elt))
290 props))
291 (setcdr (assq :plist attrs) props)
293 (apply 'define-charset-internal name (mapcar 'cdr attrs))))
296 (defun load-with-code-conversion (fullname file &optional noerror nomessage)
297 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
298 The file contents are decoded before evaluation if necessary.
299 If optional third arg NOERROR is non-nil,
300 report no error if FILE doesn't exist.
301 Print messages at start and end of loading unless
302 optional fourth arg NOMESSAGE is non-nil.
303 Return t if file exists."
304 (if (null (file-readable-p fullname))
305 (and (null noerror)
306 (signal 'file-error (list "Cannot open load file" file)))
307 ;; Read file with code conversion, and then eval.
308 (let* ((buffer
309 ;; We can't use `generate-new-buffer' because files.el
310 ;; is not yet loaded.
311 (get-buffer-create (generate-new-buffer-name " *load*")))
312 (load-in-progress t)
313 (source (save-match-data (string-match "\\.el\\'" fullname))))
314 (unless nomessage
315 (if source
316 (message "Loading %s (source)..." file)
317 (message "Loading %s..." file)))
318 (when purify-flag
319 (push (purecopy file) preloaded-file-list))
320 (unwind-protect
321 (let ((load-file-name fullname)
322 (set-auto-coding-for-load t)
323 (inhibit-file-name-operation nil))
324 (with-current-buffer buffer
325 ;; So that we don't get completely screwed if the
326 ;; file is encoded in some complicated character set,
327 ;; read it with real decoding, as a multibyte buffer.
328 (set-buffer-multibyte t)
329 ;; Don't let deactivate-mark remain set.
330 (let (deactivate-mark)
331 (insert-file-contents fullname))
332 ;; If the loaded file was inserted with no-conversion or
333 ;; raw-text coding system, make the buffer unibyte.
334 ;; Otherwise, eval-buffer might try to interpret random
335 ;; binary junk as multibyte characters.
336 (if (and enable-multibyte-characters
337 (or (eq (coding-system-type last-coding-system-used)
338 'raw-text)))
339 (set-buffer-multibyte nil))
340 ;; Make `kill-buffer' quiet.
341 (set-buffer-modified-p nil))
342 ;; Have the original buffer current while we eval.
343 (eval-buffer buffer nil
344 ;; This is compatible with what `load' does.
345 (if purify-flag file fullname)
346 nil t))
347 (let (kill-buffer-hook kill-buffer-query-functions)
348 (kill-buffer buffer)))
349 (do-after-load-evaluation fullname)
351 (unless (or nomessage noninteractive)
352 (if source
353 (message "Loading %s (source)...done" file)
354 (message "Loading %s...done" file)))
355 t)))
357 (defun charset-info (charset)
358 "Return a vector of information of CHARSET.
359 This function is provided for backward compatibility.
361 The elements of the vector are:
362 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
363 LEADING-CODE-BASE, LEADING-CODE-EXT,
364 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
365 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
366 PLIST.
367 where
368 CHARSET-ID is always 0.
369 BYTES is always 0.
370 DIMENSION is the number of bytes of a code-point of the charset:
371 1, 2, 3, or 4.
372 CHARS is the number of characters in a dimension:
373 94, 96, 128, or 256.
374 WIDTH is always 0.
375 DIRECTION is always 0.
376 LEADING-CODE-BASE is always 0.
377 LEADING-CODE-EXT is always 0.
378 ISO-FINAL-CHAR (character) is the final character of the
379 corresponding ISO 2022 charset. If the charset is not assigned
380 any final character, the value is -1.
381 ISO-GRAPHIC-PLANE is always 0.
382 REVERSE-CHARSET is always -1.
383 SHORT-NAME (string) is the short name to refer to the charset.
384 LONG-NAME (string) is the long name to refer to the charset
385 DESCRIPTION (string) is the description string of the charset.
386 PLIST (property list) may contain any type of information a user
387 want to put and get by functions `put-charset-property' and
388 `get-charset-property' respectively."
389 (vector 0
391 (charset-dimension charset)
392 (charset-chars charset)
397 (charset-iso-final-char charset)
400 (get-charset-property charset :short-name)
401 (get-charset-property charset :short-name)
402 (charset-description charset)
403 (charset-plist charset)))
405 ;; It is better not to use backquote in this file,
406 ;; because that makes a bootstrapping problem
407 ;; if you need to recompile all the Lisp files using interpreted code.
409 (defun charset-id (charset)
410 "Always return 0. This is provided for backward compatibility."
412 (make-obsolete 'charset-id "do not use it." "23.1")
414 (defmacro charset-bytes (charset)
415 "Always return 0. This is provided for backward compatibility."
417 (make-obsolete 'charset-bytes "do not use it." "23.1")
419 (defun get-charset-property (charset propname)
420 "Return the value of CHARSET's PROPNAME property.
421 This is the last value stored with
422 (put-charset-property CHARSET PROPNAME VALUE)."
423 (plist-get (charset-plist charset) propname))
425 (defun put-charset-property (charset propname value)
426 "Set CHARSETS's PROPNAME property to value VALUE.
427 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
428 (set-charset-plist charset
429 (plist-put (charset-plist charset) propname
430 (if (stringp value)
431 (purecopy value)
432 value))))
434 (defun charset-description (charset)
435 "Return description string of CHARSET."
436 (plist-get (charset-plist charset) :docstring))
438 (defun charset-dimension (charset)
439 "Return dimension of CHARSET."
440 (plist-get (charset-plist charset) :dimension))
442 (defun charset-chars (charset &optional dimension)
443 "Return number of characters contained in DIMENSION of CHARSET.
444 DIMENSION defaults to the first dimension."
445 (unless dimension (setq dimension 1))
446 (let ((code-space (plist-get (charset-plist charset) :code-space)))
447 (1+ (- (aref code-space (1- (* 2 dimension)))
448 (aref code-space (- (* 2 dimension) 2))))))
450 (defun charset-iso-final-char (charset)
451 "Return ISO-2022 final character of CHARSET.
452 Return -1 if charset isn't an ISO 2022 one."
453 (or (plist-get (charset-plist charset) :iso-final-char)
454 -1))
456 (defmacro charset-short-name (charset)
457 "Return short name of CHARSET."
458 (plist-get (charset-plist charset) :short-name))
460 (defmacro charset-long-name (charset)
461 "Return long name of CHARSET."
462 (plist-get (charset-plist charset) :long-name))
464 (defun charset-list ()
465 "Return list of all charsets ever defined."
466 charset-list)
467 (make-obsolete 'charset-list "use variable `charset-list'." "23.1")
470 ;;; CHARACTER
471 (define-obsolete-function-alias 'char-valid-p 'characterp "23.1")
473 (defun generic-char-p (char)
474 "Always return nil. This is provided for backward compatibility."
475 nil)
476 (make-obsolete 'generic-char-p "generic characters no longer exist." "23.1")
478 (defun make-char-internal (charset-id &optional code1 code2)
479 (let ((charset (aref emacs-mule-charset-table charset-id)))
480 (or charset
481 (error "Invalid Emacs-mule charset ID: %d" charset-id))
482 (make-char charset code1 code2)))
484 ;; Save the ASCII case table in case we need it later. Some locales
485 ;; (such as Turkish) modify the case behavior of ASCII characters,
486 ;; which can interfere with networking code that uses ASCII strings.
488 (defvar ascii-case-table
489 ;; Code copied from copy-case-table to avoid requiring case-table.el
490 (let ((tbl (copy-sequence (standard-case-table)))
491 (up (char-table-extra-slot (standard-case-table) 0)))
492 (if up (set-char-table-extra-slot tbl 0 (copy-sequence up)))
493 (set-char-table-extra-slot tbl 1 nil)
494 (set-char-table-extra-slot tbl 2 nil)
495 tbl)
496 "Case table for the ASCII character set.")
498 ;; Coding system stuff
500 ;; Coding system is a symbol that has been defined by the function
501 ;; `define-coding-system'.
503 (defconst coding-system-iso-2022-flags
504 '(long-form
505 ascii-at-eol
506 ascii-at-cntl
507 7-bit
508 locking-shift
509 single-shift
510 designation
511 revision
512 direction
513 init-at-bol
514 designate-at-bol
515 safe
516 latin-extra
517 composition
518 euc-tw-shift
519 use-roman
520 use-oldjis)
521 "List of symbols that control ISO-2022 encoder/decoder.
523 The value of the `:flags' attribute in the argument of the function
524 `define-coding-system' must be one of them.
526 If `long-form' is specified, use a long designation sequence on
527 encoding for the charsets `japanese-jisx0208-1978', `chinese-gb2312',
528 and `japanese-jisx0208'. The long designation sequence doesn't
529 conform to ISO 2022, but is used by such coding systems as
530 `compound-text'.
532 If `ascii-at-eol' is specified, designate ASCII to g0 at end of line
533 on encoding.
535 If `ascii-at-cntl' is specified, designate ASCII to g0 before control
536 codes and SPC on encoding.
538 If `7-bit' is specified, use 7-bit code only on encoding.
540 If `locking-shift' is specified, decode locking-shift code correctly
541 on decoding, and use locking-shift to invoke a graphic element on
542 encoding.
544 If `single-shift' is specified, decode single-shift code correctly on
545 decoding, and use single-shift to invoke a graphic element on encoding.
547 If `designation' is specified, decode designation code correctly on
548 decoding, and use designation to designate a charset to a graphic
549 element on encoding.
551 If `revision' is specified, produce an escape sequence to specify
552 revision number of a charset on encoding. Such an escape sequence is
553 always correctly decoded on decoding.
555 If `direction' is specified, decode ISO6429's code for specifying
556 direction correctly, and produce the code on encoding.
558 If `init-at-bol' is specified, on encoding, it is assumed that
559 invocation and designation statuses are reset at each beginning of
560 line even if `ascii-at-eol' is not specified; thus no codes for
561 resetting them are produced.
563 If `safe' is specified, on encoding, characters not supported by a
564 coding are replaced with `?'.
566 If `latin-extra' is specified, the code-detection routine assumes that a
567 code specified in `latin-extra-code-table' (which see) is valid.
569 If `composition' is specified, an escape sequence to specify
570 composition sequence is correctly decoded on decoding, and is produced
571 on encoding.
573 If `euc-tw-shift' is specified, the EUC-TW specific shifting code is
574 correctly decoded on decoding, and is produced on encoding.
576 If `use-roman' is specified, JIS0201-1976-Roman is designated instead
577 of ASCII.
579 If `use-oldjis' is specified, JIS0208-1976 is designated instead of
580 JIS0208-1983.")
582 (defun define-coding-system (name docstring &rest props)
583 "Define NAME (a symbol) as a coding system with DOCSTRING and attributes.
584 The remaining arguments must come in pairs ATTRIBUTE VALUE. ATTRIBUTE
585 may be any symbol.
587 The following attributes have special meanings. Those labeled as
588 \"(required)\" should not be omitted.
590 `:mnemonic' (required)
592 VALUE is a character to display on mode line for the coding system.
594 `:coding-type' (required)
596 VALUE must be one of `charset', `utf-8', `utf-16', `iso-2022',
597 `emacs-mule', `shift-jis', `ccl', `raw-text', `undecided'.
599 `:eol-type'
601 VALUE is the EOL (end-of-line) format of the coding system. It must be
602 one of `unix', `dos', `mac'. The symbol `unix' means Unix-like EOL
603 \(i.e. single LF), `dos' means DOS-like EOL \(i.e. sequence of CR LF),
604 and `mac' means Mac-like EOL \(i.e. single CR). If omitted, Emacs
605 detects the EOL format automatically when decoding.
607 `:charset-list'
609 VALUE must be a list of charsets supported by the coding system. On
610 encoding by the coding system, if a character belongs to multiple
611 charsets in the list, a charset that comes earlier in the list is
612 selected. If `:coding-type' is `iso-2022', VALUE may be `iso-2022',
613 which indicates that the coding system supports all ISO-2022 based
614 charsets. If `:coding-type' is `emacs-mule', VALUE may be
615 `emacs-mule', which indicates that the coding system supports all
616 charsets that have the `:emacs-mule-id' property.
618 `:ascii-compatible-p'
620 If VALUE is non-nil, the coding system decodes all 7-bit bytes into
621 the corresponding ASCII characters, and encodes all ASCII characters
622 back to the corresponding 7-bit bytes. VALUE defaults to nil.
624 `:decode-translation-table'
626 VALUE must be a translation table to use on decoding.
628 `:encode-translation-table'
630 VALUE must be a translation table to use on encoding.
632 `:post-read-conversion'
634 VALUE must be a function to call after some text is inserted and
635 decoded by the coding system itself and before any functions in
636 `after-insert-functions' are called. This function is passed one
637 argument; the number of characters in the text to convert, with
638 point at the start of the text. The function should leave point
639 the same, and return the new character count.
641 `:pre-write-conversion'
643 VALUE must be a function to call after all functions in
644 `write-region-annotate-functions' and `buffer-file-format' are
645 called, and before the text is encoded by the coding system
646 itself. This function should convert the whole text in the
647 current buffer. For backward compatibility, this function is
648 passed two arguments which can be ignored.
650 `:default-char'
652 VALUE must be a character. On encoding, a character not supported by
653 the coding system is replaced with VALUE.
655 `:for-unibyte'
657 VALUE non-nil means that visiting a file with the coding system
658 results in a unibyte buffer.
660 `:mime-charset'
662 VALUE must be a symbol whose name is that of a MIME charset converted
663 to lower case.
665 `:mime-text-unsuitable'
667 VALUE non-nil means the `:mime-charset' property names a charset which
668 is unsuitable for the top-level media type \"text\".
670 `:flags'
672 VALUE must be a list of symbols that control the ISO-2022 converter.
673 Each must be a member of the list `coding-system-iso-2022-flags'
674 \(which see). This attribute has a meaning only when `:coding-type'
675 is `iso-2022'.
677 `:designation'
679 VALUE must be a vector [G0-USAGE G1-USAGE G2-USAGE G3-USAGE].
680 GN-USAGE specifies the usage of graphic register GN as follows.
682 If it is nil, no charset can be designated to GN.
684 If it is a charset, the charset is initially designated to GN, and
685 never used by the other charsets.
687 If it is a list, the elements must be charsets, nil, 94, or 96. GN
688 can be used by all the listed charsets. If the list contains 94, any
689 iso-2022 charset whose code-space ranges are 94 long can be designated
690 to GN. If the list contains 96, any charsets whose whose ranges are
691 96 long can be designated to GN. If the first element is a charset,
692 that charset is initially designated to GN.
694 This attribute has a meaning only when `:coding-type' is `iso-2022'.
696 `:bom'
698 This attributes specifies whether the coding system uses a `byte order
699 mark'. VALUE must be nil, t, or cons of coding systems whose
700 `:coding-type' is `utf-16' or `utf-8'.
702 If the value is nil, on decoding, don't treat the first two-byte as
703 BOM, and on encoding, don't produce BOM bytes.
705 If the value is t, on decoding, skip the first two-byte as BOM, and on
706 encoding, produce BOM bytes accoding to the value of `:endian'.
708 If the value is cons, on decoding, check the first two-byte. If they
709 are 0xFE 0xFF, use the car part coding system of the value. If they
710 are 0xFF 0xFE, use the cdr part coding system of the value.
711 Otherwise, treat them as bytes for a normal character. On encoding,
712 produce BOM bytes accoding to the value of `:endian'.
714 This attribute has a meaning only when `:coding-type' is `utf-16' or
715 `utf-8'.
717 `:endian'
719 VALUE must be `big' or `little' specifying big-endian and
720 little-endian respectively. The default value is `big'.
722 This attribute has a meaning only when `:coding-type' is `utf-16'.
724 `:ccl-decoder'
726 VALUE is a symbol representing the registered CCL program used for
727 decoding. This attribute has a meaning only when `:coding-type' is
728 `ccl'.
730 `:ccl-encoder'
732 VALUE is a symbol representing the registered CCL program used for
733 encoding. This attribute has a meaning only when `:coding-type' is
734 `ccl'."
735 (let* ((common-attrs (mapcar 'list
736 '(:mnemonic
737 :coding-type
738 :charset-list
739 :ascii-compatible-p
740 :decode-translation-table
741 :encode-translation-table
742 :post-read-conversion
743 :pre-write-conversion
744 :default-char
745 :for-unibyte
746 :plist
747 :eol-type)))
748 (coding-type (plist-get props :coding-type))
749 (spec-attrs (mapcar 'list
750 (cond ((eq coding-type 'iso-2022)
751 '(:initial
752 :reg-usage
753 :request
754 :flags))
755 ((eq coding-type 'utf-8)
756 '(:bom))
757 ((eq coding-type 'utf-16)
758 '(:bom
759 :endian))
760 ((eq coding-type 'ccl)
761 '(:ccl-decoder
762 :ccl-encoder
763 :valids))))))
765 (dolist (slot common-attrs)
766 (setcdr slot (plist-get props (car slot))))
768 (dolist (slot spec-attrs)
769 (setcdr slot (plist-get props (car slot))))
771 (if (eq coding-type 'iso-2022)
772 (let ((designation (plist-get props :designation))
773 (flags (plist-get props :flags))
774 (initial (make-vector 4 nil))
775 (reg-usage (cons 4 4))
776 request elt)
777 (dotimes (i 4)
778 (setq elt (aref designation i))
779 (cond ((charsetp elt)
780 (aset initial i elt)
781 (setq request (cons (cons elt i) request)))
782 ((consp elt)
783 (aset initial i (car elt))
784 (if (charsetp (car elt))
785 (setq request (cons (cons (car elt) i) request)))
786 (dolist (e (cdr elt))
787 (cond ((charsetp e)
788 (setq request (cons (cons e i) request)))
789 ((eq e 94)
790 (setcar reg-usage i))
791 ((eq e 96)
792 (setcdr reg-usage i))
793 ((eq e t)
794 (setcar reg-usage i)
795 (setcdr reg-usage i)))))))
796 (setcdr (assq :initial spec-attrs) initial)
797 (setcdr (assq :reg-usage spec-attrs) reg-usage)
798 (setcdr (assq :request spec-attrs) request)
800 ;; Change :flags value from a list to a bit-mask.
801 (let ((bits 0)
802 (i 0))
803 (dolist (elt coding-system-iso-2022-flags)
804 (if (memq elt flags)
805 (setq bits (logior bits (lsh 1 i))))
806 (setq i (1+ i)))
807 (setcdr (assq :flags spec-attrs) bits))))
809 ;; Add :name and :docstring properties to PROPS.
810 (setq props
811 (cons :name (cons name (cons :docstring (cons (purecopy docstring)
812 props)))))
813 (setcdr (assq :plist common-attrs) props)
814 (apply 'define-coding-system-internal
815 name (mapcar 'cdr (append common-attrs spec-attrs)))))
817 (defun coding-system-doc-string (coding-system)
818 "Return the documentation string for CODING-SYSTEM."
819 (plist-get (coding-system-plist coding-system) :docstring))
821 (defun coding-system-mnemonic (coding-system)
822 "Return the mnemonic character of CODING-SYSTEM.
823 The mnemonic character of a coding system is used in mode line to
824 indicate the coding system. If CODING-SYSTEM is nil, return ?=."
825 (plist-get (coding-system-plist coding-system) :mnemonic))
827 (defun coding-system-type (coding-system)
828 "Return the coding type of CODING-SYSTEM.
829 A coding type is a symbol indicating the encoding method of CODING-SYSTEM.
830 See the function `define-coding-system' for more detail."
831 (plist-get (coding-system-plist coding-system) :coding-type))
833 (defun coding-system-charset-list (coding-system)
834 "Return list of charsets supported by CODING-SYSTEM.
835 If CODING-SYSTEM supports all ISO-2022 charsets, return `iso-2022'.
836 If CODING-SYSTEM supports all emacs-mule charsets, return `emacs-mule'."
837 (plist-get (coding-system-plist coding-system) :charset-list))
839 (defun coding-system-category (coding-system)
840 "Return a category symbol of CODING-SYSTEM."
841 (plist-get (coding-system-plist coding-system) :category))
843 (defun coding-system-get (coding-system prop)
844 "Extract a value from CODING-SYSTEM's property list for property PROP.
845 For compatibility with Emacs 20/21, this accepts old-style symbols
846 like `mime-charset' as well as the current style like `:mime-charset'."
847 (or (plist-get (coding-system-plist coding-system) prop)
848 (if (not (keywordp prop))
849 ;; For backward compatibility.
850 (if (eq prop 'ascii-incompatible)
851 (not (plist-get (coding-system-plist coding-system)
852 :ascii-compatible-p))
853 (plist-get (coding-system-plist coding-system)
854 (intern (concat ":" (symbol-name prop))))))))
856 (defun coding-system-eol-type-mnemonic (coding-system)
857 "Return the string indicating end-of-line format of CODING-SYSTEM."
858 (let* ((eol-type (coding-system-eol-type coding-system))
859 (val (cond ((eq eol-type 0) eol-mnemonic-unix)
860 ((eq eol-type 1) eol-mnemonic-dos)
861 ((eq eol-type 2) eol-mnemonic-mac)
862 (t eol-mnemonic-undecided))))
863 (if (stringp val)
865 (char-to-string val))))
867 (defun coding-system-lessp (x y)
868 (cond ((eq x 'no-conversion) t)
869 ((eq y 'no-conversion) nil)
870 ((eq x 'emacs-mule) t)
871 ((eq y 'emacs-mule) nil)
872 ((eq x 'undecided) t)
873 ((eq y 'undecided) nil)
874 (t (let ((c1 (coding-system-mnemonic x))
875 (c2 (coding-system-mnemonic y)))
876 (or (< (downcase c1) (downcase c2))
877 (and (not (> (downcase c1) (downcase c2)))
878 (< c1 c2)))))))
880 (defun coding-system-equal (coding-system-1 coding-system-2)
881 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
882 Two coding systems are identical if both symbols are equal
883 or one is an alias of the other."
884 (or (eq coding-system-1 coding-system-2)
885 (and (equal (coding-system-plist coding-system-1)
886 (coding-system-plist coding-system-2))
887 (let ((eol-type-1 (coding-system-eol-type coding-system-1))
888 (eol-type-2 (coding-system-eol-type coding-system-2)))
889 (or (eq eol-type-1 eol-type-2)
890 (and (vectorp eol-type-1) (vectorp eol-type-2)))))))
892 (defun add-to-coding-system-list (coding-system)
893 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
894 (if (or (null coding-system-list)
895 (coding-system-lessp coding-system (car coding-system-list)))
896 (setq coding-system-list (cons coding-system coding-system-list))
897 (let ((len (length coding-system-list))
898 mid (tem coding-system-list))
899 (while (> len 1)
900 (setq mid (nthcdr (/ len 2) tem))
901 (if (coding-system-lessp (car mid) coding-system)
902 (setq tem mid
903 len (- len (/ len 2)))
904 (setq len (/ len 2))))
905 (setcdr tem (cons coding-system (cdr tem))))))
907 (defun coding-system-list (&optional base-only)
908 "Return a list of all existing non-subsidiary coding systems.
909 If optional arg BASE-ONLY is non-nil, only base coding systems are
910 listed. The value doesn't include subsidiary coding systems which are
911 made from bases and aliases automatically for various end-of-line
912 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
913 (let ((codings nil))
914 (dolist (coding coding-system-list)
915 (if (eq (coding-system-base coding) coding)
916 (if base-only
917 (setq codings (cons coding codings))
918 (dolist (alias (coding-system-aliases coding))
919 (setq codings (cons alias codings))))))
920 codings))
922 (defconst char-coding-system-table nil
923 "It exists just for backward compatibility, and the value is always nil.")
924 (make-obsolete-variable 'char-coding-system-table nil "23.1")
926 (defun transform-make-coding-system-args (name type &optional doc-string props)
927 "For internal use only.
928 Transform XEmacs style args for `make-coding-system' to Emacs style.
929 Value is a list of transformed arguments."
930 (let ((mnemonic (string-to-char (or (plist-get props 'mnemonic) "?")))
931 (eol-type (plist-get props 'eol-type))
932 properties tmp)
933 (cond
934 ((eq eol-type 'lf) (setq eol-type 'unix))
935 ((eq eol-type 'crlf) (setq eol-type 'dos))
936 ((eq eol-type 'cr) (setq eol-type 'mac)))
937 (if (setq tmp (plist-get props 'post-read-conversion))
938 (setq properties (plist-put properties 'post-read-conversion tmp)))
939 (if (setq tmp (plist-get props 'pre-write-conversion))
940 (setq properties (plist-put properties 'pre-write-conversion tmp)))
941 (cond
942 ((eq type 'shift-jis)
943 `(,name 1 ,mnemonic ,doc-string () ,properties ,eol-type))
944 ((eq type 'iso2022) ; This is not perfect.
945 (if (plist-get props 'escape-quoted)
946 (error "escape-quoted is not supported: %S"
947 `(,name ,type ,doc-string ,props)))
948 (let ((g0 (plist-get props 'charset-g0))
949 (g1 (plist-get props 'charset-g1))
950 (g2 (plist-get props 'charset-g2))
951 (g3 (plist-get props 'charset-g3))
952 (use-roman
953 (and
954 (eq (cadr (assoc 'latin-jisx0201
955 (plist-get props 'input-charset-conversion)))
956 'ascii)
957 (eq (cadr (assoc 'ascii
958 (plist-get props 'output-charset-conversion)))
959 'latin-jisx0201)))
960 (use-oldjis
961 (and
962 (eq (cadr (assoc 'japanese-jisx0208-1978
963 (plist-get props 'input-charset-conversion)))
964 'japanese-jisx0208)
965 (eq (cadr (assoc 'japanese-jisx0208
966 (plist-get props 'output-charset-conversion)))
967 'japanese-jisx0208-1978))))
968 (if (charsetp g0)
969 (if (plist-get props 'force-g0-on-output)
970 (setq g0 `(nil ,g0))
971 (setq g0 `(,g0 t))))
972 (if (charsetp g1)
973 (if (plist-get props 'force-g1-on-output)
974 (setq g1 `(nil ,g1))
975 (setq g1 `(,g1 t))))
976 (if (charsetp g2)
977 (if (plist-get props 'force-g2-on-output)
978 (setq g2 `(nil ,g2))
979 (setq g2 `(,g2 t))))
980 (if (charsetp g3)
981 (if (plist-get props 'force-g3-on-output)
982 (setq g3 `(nil ,g3))
983 (setq g3 `(,g3 t))))
984 `(,name 2 ,mnemonic ,doc-string
985 (,g0 ,g1 ,g2 ,g3
986 ,(plist-get props 'short)
987 ,(not (plist-get props 'no-ascii-eol))
988 ,(not (plist-get props 'no-ascii-cntl))
989 ,(plist-get props 'seven)
991 ,(not (plist-get props 'lock-shift))
992 ,use-roman
993 ,use-oldjis
994 ,(plist-get props 'no-iso6429)
995 nil nil nil nil)
996 ,properties ,eol-type)))
997 ((eq type 'big5)
998 `(,name 3 ,mnemonic ,doc-string () ,properties ,eol-type))
999 ((eq type 'ccl)
1000 `(,name 4 ,mnemonic ,doc-string
1001 (,(plist-get props 'decode) . ,(plist-get props 'encode))
1002 ,properties ,eol-type))
1004 (error "unsupported XEmacs style make-coding-style arguments: %S"
1005 `(,name ,type ,doc-string ,props))))))
1007 (defun make-coding-system (coding-system type mnemonic doc-string
1008 &optional
1009 flags
1010 properties
1011 eol-type)
1012 "Define a new coding system CODING-SYSTEM (symbol).
1013 This function is provided for backward compatibility."
1014 ;; For compatibility with XEmacs, we check the type of TYPE. If it
1015 ;; is a symbol, perhaps, this function is called with XEmacs-style
1016 ;; arguments. Here, try to transform that kind of arguments to
1017 ;; Emacs style.
1018 (if (symbolp type)
1019 (let ((args (transform-make-coding-system-args coding-system type
1020 mnemonic doc-string)))
1021 (setq coding-system (car args)
1022 type (nth 1 args)
1023 mnemonic (nth 2 args)
1024 doc-string (nth 3 args)
1025 flags (nth 4 args)
1026 properties (nth 5 args)
1027 eol-type (nth 6 args))))
1029 (setq type
1030 (cond ((eq type 0) 'emacs-mule)
1031 ((eq type 1) 'shift-jis)
1032 ((eq type 2) 'iso2022)
1033 ((eq type 3) 'big5)
1034 ((eq type 4) 'ccl)
1035 ((eq type 5) 'raw-text)
1037 (error "Invalid coding system type: %s" type))))
1039 (setq properties
1040 (let ((plist nil) key)
1041 (dolist (elt properties)
1042 (setq key (car elt))
1043 (cond ((eq key 'post-read-conversion)
1044 (setq key :post-read-conversion))
1045 ((eq key 'pre-write-conversion)
1046 (setq key :pre-write-conversion))
1047 ((eq key 'translation-table-for-decode)
1048 (setq key :decode-translation-table))
1049 ((eq key 'translation-table-for-encode)
1050 (setq key :encode-translation-table))
1051 ((eq key 'safe-charsets)
1052 (setq key :charset-list))
1053 ((eq key 'mime-charset)
1054 (setq key :mime-charset))
1055 ((eq key 'valid-codes)
1056 (setq key :valids)))
1057 (setq plist (plist-put plist key (cdr elt))))
1058 plist))
1059 (setq properties (plist-put properties :mnemonic mnemonic))
1060 (plist-put properties :coding-type type)
1061 (cond ((eq eol-type 0) (setq eol-type 'unix))
1062 ((eq eol-type 1) (setq eol-type 'dos))
1063 ((eq eol-type 2) (setq eol-type 'mac))
1064 ((vectorp eol-type) (setq eol-type nil)))
1065 (plist-put properties :eol-type eol-type)
1067 (cond
1068 ((eq type 'iso2022)
1069 (plist-put properties :flags
1070 (list (and (or (consp (nth 0 flags))
1071 (consp (nth 1 flags))
1072 (consp (nth 2 flags))
1073 (consp (nth 3 flags))) 'designation)
1074 (or (nth 4 flags) 'long-form)
1075 (and (nth 5 flags) 'ascii-at-eol)
1076 (and (nth 6 flags) 'ascii-at-cntl)
1077 (and (nth 7 flags) '7-bit)
1078 (and (nth 8 flags) 'locking-shift)
1079 (and (nth 9 flags) 'single-shift)
1080 (and (nth 10 flags) 'use-roman)
1081 (and (nth 11 flags) 'use-oldjis)
1082 (or (nth 12 flags) 'direction)
1083 (and (nth 13 flags) 'init-at-bol)
1084 (and (nth 14 flags) 'designate-at-bol)
1085 (and (nth 15 flags) 'safe)
1086 (and (nth 16 flags) 'latin-extra)))
1087 (plist-put properties :designation
1088 (let ((vec (make-vector 4 nil)))
1089 (dotimes (i 4)
1090 (let ((spec (nth i flags)))
1091 (if (eq spec t)
1092 (aset vec i '(94 96))
1093 (if (consp spec)
1094 (progn
1095 (if (memq t spec)
1096 (setq spec (append (delq t spec) '(94 96))))
1097 (aset vec i spec))))))
1098 vec)))
1100 ((eq type 'ccl)
1101 (plist-put properties :ccl-decoder (car flags))
1102 (plist-put properties :ccl-encoder (cdr flags))))
1104 (apply 'define-coding-system coding-system doc-string properties))
1106 (make-obsolete 'make-coding-system 'define-coding-system "23.1")
1108 (defun merge-coding-systems (first second)
1109 "Fill in any unspecified aspects of coding system FIRST from SECOND.
1110 Return the resulting coding system."
1111 (let ((base (coding-system-base second))
1112 (eol (coding-system-eol-type second)))
1113 ;; If FIRST doesn't specify text conversion, merge with that of SECOND.
1114 (if (eq (coding-system-base first) 'undecided)
1115 (setq first (coding-system-change-text-conversion first base)))
1116 ;; If FIRST doesn't specify eol conversion, merge with that of SECOND.
1117 (if (and (vectorp (coding-system-eol-type first))
1118 (numberp eol) (>= eol 0) (<= eol 2))
1119 (setq first (coding-system-change-eol-conversion
1120 first eol)))
1121 first))
1123 (defun autoload-coding-system (symbol form)
1124 "Define SYMBOL as a coding-system that is defined on demand.
1126 FORM is a form to evaluate to define the coding-system."
1127 (put symbol 'coding-system-define-form form)
1128 (setq coding-system-alist (cons (list (symbol-name symbol))
1129 coding-system-alist))
1130 (dolist (elt '("-unix" "-dos" "-mac"))
1131 (let ((name (concat (symbol-name symbol) elt)))
1132 (put (intern name) 'coding-system-define-form form)
1133 (setq coding-system-alist (cons (list name) coding-system-alist)))))
1135 ;; This variable is set in these three cases:
1136 ;; (1) A file is read by a coding system specified explicitly.
1137 ;; after-insert-file-set-coding sets the car of this value to
1138 ;; coding-system-for-read, and sets the cdr to nil.
1139 ;; (2) A buffer is saved.
1140 ;; After writing, basic-save-buffer-1 sets the car of this value
1141 ;; to last-coding-system-used.
1142 ;; (3) set-buffer-file-coding-system is called.
1143 ;; The cdr of this value is set to the specified coding system.
1144 ;; This variable is used for decoding in revert-buffer and encoding in
1145 ;; select-safe-coding-system.
1146 (defvar buffer-file-coding-system-explicit nil
1147 "The file coding system explicitly specified for the current buffer.
1148 The value is a cons of coding systems for reading (decoding) and
1149 writing (encoding).
1150 Internal use only.")
1151 (make-variable-buffer-local 'buffer-file-coding-system-explicit)
1152 (put 'buffer-file-coding-system-explicit 'permanent-local t)
1154 (defun read-buffer-file-coding-system ()
1155 (let* ((bcss (find-coding-systems-region (point-min) (point-max)))
1156 (css-table
1157 (unless (equal bcss '(undecided))
1158 (append '("dos" "unix" "mac")
1159 (delq nil (mapcar (lambda (cs)
1160 (if (memq (coding-system-base cs) bcss)
1161 (symbol-name cs)))
1162 coding-system-list)))))
1163 (combined-table
1164 (if css-table
1165 (completion-table-in-turn css-table coding-system-alist)
1166 coding-system-alist))
1167 (auto-cs
1168 (unless find-file-literally
1169 (save-excursion
1170 (save-restriction
1171 (widen)
1172 (goto-char (point-min))
1173 (funcall set-auto-coding-function
1174 (or buffer-file-name "") (buffer-size))))))
1175 (preferred
1176 (let ((bfcs (default-value 'buffer-file-coding-system)))
1177 (cons (and (or (equal bcss '(undecided))
1178 (memq (coding-system-base bfcs) bcss))
1179 bfcs)
1180 (mapcar (lambda (cs)
1181 (and (coding-system-p cs)
1182 (coding-system-get cs :mime-charset)
1183 (or (equal bcss '(undecided))
1184 (memq (coding-system-base cs) bcss))
1185 cs))
1186 (coding-system-priority-list)))))
1187 (default
1188 (let ((current (coding-system-base buffer-file-coding-system)))
1189 ;; Generally use as a default the first preferred coding-system
1190 ;; different from the current coding-system, except for
1191 ;; the case of auto-cs since choosing anything else is asking
1192 ;; for trouble (would lead to using a different coding
1193 ;; system than specified in the coding tag).
1194 (or auto-cs
1195 (car (delq nil
1196 (mapcar (lambda (cs)
1197 (if (eq current (coding-system-base cs))
1199 cs))
1200 preferred))))))
1201 (completion-ignore-case t)
1202 (completion-pcm--delim-wild-regex ; Let "u8" complete to "utf-8".
1203 (concat completion-pcm--delim-wild-regex
1204 "\\|\\([[:alpha:]]\\)[[:digit:]]"))
1205 (cs (completing-read
1206 (format "Coding system for saving file (default %s): " default)
1207 combined-table
1208 nil t nil 'coding-system-history
1209 (if default (symbol-name default)))))
1210 (unless (zerop (length cs)) (intern cs))))
1212 (defun set-buffer-file-coding-system (coding-system &optional force nomodify)
1213 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1214 This means that when you save the buffer, it will be converted
1215 according to CODING-SYSTEM. For a list of possible values of
1216 CODING-SYSTEM, use \\[list-coding-systems].
1218 If CODING-SYSTEM leaves the text conversion unspecified, or if it leaves
1219 the end-of-line conversion unspecified, FORCE controls what to do.
1220 If FORCE is nil, get the unspecified aspect (or aspects) from the buffer's
1221 previous `buffer-file-coding-system' value (if it is specified there).
1222 Otherwise, leave it unspecified.
1224 This marks the buffer modified so that the succeeding \\[save-buffer]
1225 surely saves the buffer with CODING-SYSTEM. From a program, if you
1226 don't want to mark the buffer modified, specify t for NOMODIFY.
1227 If you know exactly what coding system you want to use,
1228 just set the variable `buffer-file-coding-system' directly."
1229 (interactive
1230 (list (read-buffer-file-coding-system)
1231 current-prefix-arg))
1232 (check-coding-system coding-system)
1233 (if (and coding-system buffer-file-coding-system (null force))
1234 (setq coding-system
1235 (merge-coding-systems coding-system buffer-file-coding-system)))
1236 (when (called-interactively-p 'interactive)
1237 ;; Check whether save would succeed, and jump to the offending char(s)
1238 ;; if not.
1239 (let ((css (find-coding-systems-region (point-min) (point-max))))
1240 (unless (or (eq (car css) 'undecided)
1241 (memq (coding-system-base coding-system) css))
1242 (setq coding-system (select-safe-coding-system-interactively
1243 (point-min) (point-max) css
1244 (list coding-system))))))
1245 (setq buffer-file-coding-system coding-system)
1246 (if buffer-file-coding-system-explicit
1247 (setcdr buffer-file-coding-system-explicit coding-system)
1248 (setq buffer-file-coding-system-explicit (cons nil coding-system)))
1249 (unless nomodify
1250 (set-buffer-modified-p t))
1251 (force-mode-line-update))
1253 (defun revert-buffer-with-coding-system (coding-system &optional force)
1254 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1255 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1257 If CODING-SYSTEM leaves the text conversion unspecified, or if it leaves
1258 the end-of-line conversion unspecified, FORCE controls what to do.
1259 If FORCE is nil, get the unspecified aspect (or aspects) from the buffer's
1260 previous `buffer-file-coding-system' value (if it is specified there).
1261 Otherwise, determine it from the file contents as usual for visiting a file."
1262 (interactive "zCoding system for visited file (default nil): \nP")
1263 (check-coding-system coding-system)
1264 (if (and coding-system buffer-file-coding-system (null force))
1265 (setq coding-system
1266 (merge-coding-systems coding-system buffer-file-coding-system)))
1267 (let ((coding-system-for-read coding-system))
1268 (revert-buffer)))
1270 (defun set-file-name-coding-system (coding-system)
1271 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1272 It actually just set the variable `file-name-coding-system' (which see)
1273 to CODING-SYSTEM."
1274 (interactive "zCoding system for file names (default nil): ")
1275 (check-coding-system coding-system)
1276 (if (and coding-system
1277 (not (coding-system-get coding-system :ascii-compatible-p))
1278 (not (coding-system-get coding-system :suitable-for-file-name)))
1279 (error "%s is not suitable for file names" coding-system))
1280 (setq file-name-coding-system coding-system))
1282 (defvar default-terminal-coding-system nil
1283 "Default value for the terminal coding system.
1284 This is normally set according to the selected language environment.
1285 See also the command `set-terminal-coding-system'.")
1287 (defun set-terminal-coding-system (coding-system &optional terminal)
1288 "Set coding system of terminal output to CODING-SYSTEM.
1289 All text output to TERMINAL will be encoded
1290 with the specified coding system.
1292 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1293 The default is determined by the selected language environment
1294 or by the previous use of this command.
1296 TERMINAL may be a terminal object, a frame, or nil for the
1297 selected frame's terminal. The setting has no effect on
1298 graphical terminals."
1299 (interactive
1300 (list (let ((default (if (and (not (terminal-coding-system))
1301 default-terminal-coding-system)
1302 default-terminal-coding-system)))
1303 (read-coding-system
1304 (format "Coding system for terminal display (default %s): "
1305 default)
1306 default))))
1307 (if (and (not coding-system)
1308 (not (terminal-coding-system)))
1309 (setq coding-system default-terminal-coding-system))
1310 (if coding-system
1311 (setq default-terminal-coding-system coding-system))
1312 (set-terminal-coding-system-internal coding-system terminal)
1313 (redraw-frame (selected-frame)))
1315 (defvar default-keyboard-coding-system nil
1316 "Default value of the keyboard coding system.
1317 This is normally set according to the selected language environment.
1318 See also the command `set-keyboard-coding-system'.")
1320 (defun set-keyboard-coding-system (coding-system &optional terminal)
1321 "Set coding system for keyboard input on TERMINAL to CODING-SYSTEM.
1323 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1324 The default is determined by the selected language environment
1325 or by the previous use of this command.
1327 If CODING-SYSTEM is nil or the coding-type of CODING-SYSTEM is
1328 `raw-text', the decoding of keyboard input is disabled.
1330 TERMINAL may be a terminal object, a frame, or nil for the
1331 selected frame's terminal. The setting has no effect on
1332 graphical terminals."
1333 (interactive
1334 (list (let* ((coding (keyboard-coding-system nil))
1335 (default (if (eq (coding-system-type coding) 'raw-text)
1336 default-keyboard-coding-system)))
1337 (read-coding-system
1338 (format "Coding system for keyboard input (default %s): "
1339 default)
1340 default))))
1341 (let ((coding-type (coding-system-type coding-system))
1342 (saved-meta-mode
1343 (terminal-parameter terminal 'keyboard-coding-saved-meta-mode)))
1344 (if (not (eq coding-type 'raw-text))
1345 (let (accept-8-bit)
1346 (if (not (or (coding-system-get coding-system :suitable-for-keyboard)
1347 (coding-system-get coding-system :ascii-compatible-p)))
1348 (error "Unsuitable coding system for keyboard: %s" coding-system))
1349 (cond ((memq coding-type '(charset utf-8 shift-jis big5 ccl))
1350 (setq accept-8-bit t))
1351 ((eq coding-type 'iso-2022)
1352 (let ((flags (coding-system-get coding-system :flags)))
1353 (or (memq '7-bit flags)
1354 (setq accept-8-bit t))))
1356 (error "Unsupported coding system for keyboard: %s"
1357 coding-system)))
1358 (when accept-8-bit
1359 (or saved-meta-mode
1360 (set-terminal-parameter terminal
1361 'keyboard-coding-saved-meta-mode
1362 (cons (nth 2 (current-input-mode))
1363 nil)))
1364 (set-input-meta-mode 8))
1365 ;; Avoid end-of-line conversion.
1366 (setq coding-system
1367 (coding-system-change-eol-conversion coding-system 'unix)))
1369 (when saved-meta-mode
1370 (set-input-meta-mode (car saved-meta-mode))
1371 (set-terminal-parameter terminal
1372 'keyboard-coding-saved-meta-mode
1373 nil))))
1374 (set-keyboard-coding-system-internal coding-system terminal)
1375 (setq keyboard-coding-system coding-system))
1377 (defcustom keyboard-coding-system nil
1378 "Specify coding system for keyboard input.
1379 If you set this on a terminal which can't distinguish Meta keys from
1380 8-bit characters, you will have to use ESC to type Meta characters.
1381 See Info node `Terminal Coding' and Info node `Unibyte Mode'.
1383 On non-windowing terminals, this is set from the locale by default.
1385 Setting this variable directly does not take effect;
1386 use either \\[customize] or \\[set-keyboard-coding-system]."
1387 :type '(coding-system :tag "Coding system")
1388 :link '(info-link "(emacs)Terminal Coding")
1389 :link '(info-link "(emacs)Unibyte Mode")
1390 :set (lambda (symbol value)
1391 ;; Don't load encoded-kb unnecessarily.
1392 (if (or value (boundp 'encoded-kbd-setup-display))
1393 (set-keyboard-coding-system value)
1394 (set-default 'keyboard-coding-system nil))) ; must initialize
1395 :version "22.1"
1396 :group 'keyboard
1397 :group 'mule)
1399 (defun set-buffer-process-coding-system (decoding encoding)
1400 "Set coding systems for the process associated with the current buffer.
1401 DECODING is the coding system to be used to decode input from the process,
1402 ENCODING is the coding system to be used to encode output to the process.
1404 For a list of possible coding systems, use \\[list-coding-systems]."
1405 (interactive
1406 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1407 (let ((proc (get-buffer-process (current-buffer))))
1408 (if (null proc)
1409 (error "No process")
1410 (check-coding-system decoding)
1411 (check-coding-system encoding)
1412 (set-process-coding-system proc decoding encoding)))
1413 (force-mode-line-update))
1415 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
1417 (defun set-selection-coding-system (coding-system)
1418 "Make CODING-SYSTEM used for communicating with other X clients.
1419 When sending or receiving text via cut_buffer, selection, and clipboard,
1420 the text is encoded or decoded by CODING-SYSTEM."
1421 (interactive "zCoding system for X selection: ")
1422 (check-coding-system coding-system)
1423 (setq selection-coding-system coding-system))
1425 ;; Coding system lastly specified by the command
1426 ;; set-next-selection-coding-system.
1427 (defvar last-next-selection-coding-system nil)
1429 (defun set-next-selection-coding-system (coding-system)
1430 "Use CODING-SYSTEM for next communication with other window system clients.
1431 This setting is effective for the next communication only."
1432 (interactive
1433 (list (read-coding-system
1434 (if last-next-selection-coding-system
1435 (format "Coding system for the next selection (default %S): "
1436 last-next-selection-coding-system)
1437 "Coding system for the next selection: ")
1438 last-next-selection-coding-system)))
1439 (if coding-system
1440 (setq last-next-selection-coding-system coding-system)
1441 (setq coding-system last-next-selection-coding-system))
1442 (check-coding-system coding-system)
1444 (setq next-selection-coding-system coding-system))
1446 (defun set-coding-priority (arg)
1447 "Set priority of coding categories according to ARG.
1448 ARG is a list of coding categories ordered by priority.
1450 This function is provided for backward compatibility."
1451 (apply 'set-coding-system-priority
1452 (mapcar #'(lambda (x) (symbol-value x)) arg)))
1453 (make-obsolete 'set-coding-priority 'set-coding-system-priority "23.1")
1455 ;;; X selections
1457 (defvar ctext-non-standard-encodings-alist
1458 (mapcar 'purecopy
1459 '(("big5-0" big5 2 big5)
1460 ("ISO8859-14" iso-8859-14 1 latin-iso8859-14)
1461 ("ISO8859-15" iso-8859-15 1 latin-iso8859-15)
1462 ("gbk-0" gbk 2 chinese-gbk)
1463 ("koi8-r" koi8-r 1 koi8-r)
1464 ("microsoft-cp1251" windows-1251 1 windows-1251)))
1465 "Alist of non-standard encoding names vs the corresponding usages in CTEXT.
1467 It controls how extended segments of a compound text are handled
1468 by the coding system `compound-text-with-extensions'.
1470 Each element has the form (ENCODING-NAME CODING-SYSTEM N-OCTET CHARSET).
1472 ENCODING-NAME is an encoding name of an \"extended segment\".
1474 CODING-SYSTEM is the coding-system to encode (or decode) the
1475 characters into (or from) the extended segment.
1477 N-OCTET is the number of octets (bytes) that encodes a character
1478 in the segment. It can be 0 (meaning the number of octets per
1479 character is variable), 1, 2, 3, or 4.
1481 CHARSET is a character set containing characters that are encoded
1482 in the segment. It can be a list of character sets.
1484 On decoding CTEXT, all encoding names listed here are recognized.
1486 On encoding CTEXT, encoding names in the variable
1487 `ctext-non-standard-encodings' (which see) and in the information
1488 listed for the current language environment under the key
1489 `ctext-non-standard-encodings' are used.")
1491 (defvar ctext-non-standard-encodings nil
1492 "List of non-standard encoding names used in extended segments of CTEXT.
1493 Each element must be one of the names listed in the variable
1494 `ctext-non-standard-encodings-alist' (which see).")
1496 (defvar ctext-non-standard-encodings-regexp
1497 (purecopy
1498 (string-to-multibyte
1499 (concat
1500 ;; For non-standard encodings.
1501 "\\(\e%/[0-4][\200-\377][\200-\377]\\([^\002]+\\)\002\\)"
1502 "\\|"
1503 ;; For UTF-8 encoding.
1504 "\\(\e%G[^\e]*\e%@\\)"))))
1506 ;; Functions to support "Non-Standard Character Set Encodings" defined
1507 ;; by the COMPOUND-TEXT spec. They also support "The UTF-8 encoding"
1508 ;; described in the section 7 of the documentation of COMPOUND-TEXT
1509 ;; distributed with XFree86.
1511 (defun ctext-post-read-conversion (len)
1512 "Decode LEN characters encoded as Compound Text with Extended Segments."
1513 ;; We don't need the following because it is expected that this
1514 ;; function is mainly used for decoding X selection which is not
1515 ;; that big data.
1516 ;;(buffer-disable-undo) ; minimize consing due to insertions and deletions
1517 (save-match-data
1518 (save-restriction
1519 (narrow-to-region (point) (+ (point) len))
1520 (let ((case-fold-search nil)
1521 last-coding-system-used
1522 pos bytes)
1523 (decode-coding-region (point-min) (point-max) 'ctext)
1524 (while (re-search-forward ctext-non-standard-encodings-regexp
1525 nil 'move)
1526 (setq pos (match-beginning 0))
1527 (if (match-beginning 1)
1528 ;; ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
1529 (let* ((M (multibyte-char-to-unibyte (char-after (+ pos 4))))
1530 (L (multibyte-char-to-unibyte (char-after (+ pos 5))))
1531 (encoding (match-string 2))
1532 (encoding-info (assoc-string
1533 encoding
1534 ctext-non-standard-encodings-alist t))
1535 (coding (if encoding-info
1536 (nth 1 encoding-info)
1537 (setq encoding (intern (downcase encoding)))
1538 (and (coding-system-p encoding)
1539 encoding))))
1540 (setq bytes (- (+ (* (- M 128) 128) (- L 128))
1541 (- (point) (+ pos 6))))
1542 (when coding
1543 (delete-region pos (point))
1544 (forward-char bytes)
1545 (decode-coding-region (- (point) bytes) (point) coding)))
1546 ;; ESC % G --UTF-8-BYTES-- ESC % @
1547 (delete-char -3)
1548 (delete-region pos (+ pos 3))
1549 (decode-coding-region pos (point) 'utf-8))))
1550 (goto-char (point-min))
1551 (- (point-max) (point)))))
1553 (defvar ctext-standard-encodings
1554 '(ascii latin-jisx0201 katakana-jisx0201
1555 latin-iso8859-1 latin-iso8859-2 latin-iso8859-3 latin-iso8859-4
1556 greek-iso8859-7 arabic-iso8859-6 hebrew-iso8859-8 cyrillic-iso8859-5
1557 latin-iso8859-9
1558 chinese-gb2312 japanese-jisx0208 korean-ksc5601)
1559 "List of approved standard encodings (i.e. charsets) of X's Compound Text.
1560 Coding-system `compound-text-with-extensions' encodes a character
1561 belonging to any of those charsets using the normal ISO2022
1562 designation sequence unless the current language environment or
1563 the variable `ctext-non-standard-encodings' decide to use an extended
1564 segment of CTEXT for that character. See also the documentation
1565 of `ctext-non-standard-encodings-alist'.")
1567 ;; Return an alist of CHARSET vs CTEXT-USAGE-INFO generated from
1568 ;; `ctext-non-standard-encodings' and a list specified by the key
1569 ;; `ctext-non-standard-encodings' for the currrent language
1570 ;; environment. CTEXT-USAGE-INFO is one of the element of
1571 ;; `ctext-non-standard-encodings-alist' or nil. In the former case, a
1572 ;; character in CHARSET is encoded using extended segment. In the
1573 ;; latter case, a character in CHARSET is encoded using normal ISO2022
1574 ;; designation sequence. If a character is not in any of CHARSETs, it
1575 ;; is encoded using UTF-8 encoding extention.
1577 (defun ctext-non-standard-encodings-table ()
1578 (let* ((table (append ctext-non-standard-encodings
1579 (copy-sequence
1580 (get-language-info current-language-environment
1581 'ctext-non-standard-encodings))))
1582 (tail table)
1583 elt)
1584 (while tail
1585 (setq elt (car tail))
1586 (let* ((slot (assoc elt ctext-non-standard-encodings-alist))
1587 (charset (nth 3 slot)))
1588 (if (charsetp charset)
1589 (setcar tail
1590 (cons (plist-get (charset-plist charset) :base) slot))
1591 (setcar tail (cons (car charset) slot))
1592 (dolist (cs (cdr charset))
1593 (setcdr tail
1594 (cons (cons (plist-get (charset-plist (car cs)) :base) slot)
1595 (cdr tail)))
1596 (setq tail (cdr tail))))
1597 (setq tail (cdr tail))))
1598 table))
1600 (defun ctext-pre-write-conversion (from to)
1601 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1603 If FROM is a string, generate a new temp buffer, insert the text,
1604 and convert it in the temporary buffer. Otherwise, convert
1605 in-place."
1606 (save-match-data
1607 ;; Setup a working buffer if necessary.
1608 (when (stringp from)
1609 (set-buffer (generate-new-buffer " *temp"))
1610 (set-buffer-multibyte (multibyte-string-p from))
1611 (insert from)
1612 (setq from (point-min) to (point-max)))
1613 (save-restriction
1614 (narrow-to-region from to)
1615 (goto-char from)
1616 (let ((encoding-table (ctext-non-standard-encodings-table))
1617 (charset-list (sort-charsets
1618 (copy-sequence ctext-standard-encodings)))
1619 (end-pos (make-marker))
1620 last-coding-system-used
1621 last-pos charset encoding-info)
1622 (dolist (elt encoding-table)
1623 (push (car elt) charset-list))
1624 (setq end-pos (point-marker))
1625 (while (re-search-forward "[^\0-\177]+" nil t)
1626 ;; Found a sequence of non-ASCII characters.
1627 (set-marker end-pos (match-end 0))
1628 (goto-char (match-beginning 0))
1629 (setq last-pos (point)
1630 charset (char-charset (following-char) charset-list))
1631 (forward-char 1)
1632 (while (and (< (point) end-pos)
1633 (eq charset (char-charset (following-char) charset-list)))
1634 (forward-char 1))
1635 (if charset
1636 (if (setq encoding-info (cdr (assq charset encoding-table)))
1637 ;; Encode this range using an extended segment.
1638 (let ((encoding-name (car encoding-info))
1639 (coding-system (nth 1 encoding-info))
1640 (noctets (nth 2 encoding-info))
1641 len)
1642 (encode-coding-region last-pos (point) coding-system)
1643 (setq len (+ (length encoding-name) 1
1644 (- (point) last-pos)))
1645 ;; According to the spec of CTEXT, it is not
1646 ;; necessary to produce this extra designation
1647 ;; sequence, but some buggy application
1648 ;; (e.g. crxvt-gb) requires it.
1649 (insert "\e(B")
1650 (save-excursion
1651 (goto-char last-pos)
1652 (insert (format "\e%%/%d" noctets))
1653 (insert-byte (+ (/ len 128) 128) 1)
1654 (insert-byte (+ (% len 128) 128) 1)
1655 (insert encoding-name)
1656 (insert 2)))
1657 ;; Encode this range as characters in CHARSET.
1658 (put-text-property last-pos (point) 'charset charset))
1659 ;; Encode this range using UTF-8 encoding extention.
1660 (encode-coding-region last-pos (point) 'mule-utf-8)
1661 (save-excursion
1662 (goto-char last-pos)
1663 (insert "\e%G"))
1664 (insert "\e%@")))
1665 (goto-char (point-min)))))
1666 ;; Must return nil, as build_annotations_2 expects that.
1667 nil)
1669 ;;; FILE I/O
1671 (defcustom auto-coding-alist
1672 ;; .exe and .EXE are added to support archive-mode looking at DOS
1673 ;; self-extracting exe archives.
1674 (mapcar (lambda (arg) (cons (purecopy (car arg)) (cdr arg)))
1675 '(("\\.\\(\
1676 arc\\|zip\\|lzh\\|lha\\|zoo\\|[jew]ar\\|xpi\\|rar\\|7z\\|\
1677 ARC\\|ZIP\\|LZH\\|LHA\\|ZOO\\|[JEW]AR\\|XPI\\|RAR\\|7Z\\)\\'"
1678 . no-conversion-multibyte)
1679 ("\\.\\(exe\\|EXE\\)\\'" . no-conversion)
1680 ("\\.\\(sx[dmicw]\\|odt\\|tar\\|tgz\\)\\'" . no-conversion)
1681 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|xz\\|gpg\\)\\'" . no-conversion)
1682 ("\\.\\(jpe?g\\|png\\|gif\\|tiff?\\|p[bpgn]m\\)\\'" . no-conversion)
1683 ("\\.pdf\\'" . no-conversion)
1684 ("/#[^/]+#\\'" . emacs-mule)))
1685 "Alist of filename patterns vs corresponding coding systems.
1686 Each element looks like (REGEXP . CODING-SYSTEM).
1687 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1689 The settings in this alist take priority over `coding:' tags
1690 in the file (see the function `set-auto-coding')
1691 and the contents of `file-coding-system-alist'."
1692 :version "24.1" ; added xz
1693 :group 'files
1694 :group 'mule
1695 :type '(repeat (cons (regexp :tag "File name regexp")
1696 (symbol :tag "Coding system"))))
1698 (defcustom auto-coding-regexp-alist
1699 (mapcar (lambda (arg) (cons (purecopy (car arg)) (cdr arg)))
1700 '(("\\`BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion)
1701 ("\\`\xFE\xFF" . utf-16be-with-signature)
1702 ("\\`\xFF\xFE" . utf-16le-with-signature)
1703 ("\\`\xEF\xBB\xBF" . utf-8-with-signature)
1704 ("\\`;ELC\024\0\0\0" . emacs-mule))) ; Emacs 20-compiled
1705 "Alist of patterns vs corresponding coding systems.
1706 Each element looks like (REGEXP . CODING-SYSTEM).
1707 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1709 The settings in this alist take priority over `coding:' tags
1710 in the file (see the function `set-auto-coding')
1711 and the contents of `file-coding-system-alist'."
1712 :group 'files
1713 :group 'mule
1714 :type '(repeat (cons (regexp :tag "Regexp")
1715 (symbol :tag "Coding system"))))
1717 (defun auto-coding-regexp-alist-lookup (from to)
1718 "Lookup `auto-coding-regexp-alist' for the contents of the current buffer.
1719 The value is a coding system is specified for the region FROM and TO,
1720 or nil."
1721 (save-excursion
1722 (goto-char from)
1723 (let ((alist auto-coding-regexp-alist)
1724 coding-system)
1725 (while (and alist (not coding-system))
1726 (let ((regexp (car (car alist))))
1727 (if enable-multibyte-characters
1728 (setq regexp (string-to-multibyte regexp)))
1729 (if (re-search-forward regexp to t)
1730 (setq coding-system (cdr (car alist)))
1731 (setq alist (cdr alist)))))
1732 coding-system)))
1734 ;; See the bottom of this file for built-in auto coding functions.
1735 (defcustom auto-coding-functions '(sgml-xml-auto-coding-function
1736 sgml-html-meta-auto-coding-function)
1737 "A list of functions which attempt to determine a coding system.
1739 Each function in this list should be written to operate on the
1740 current buffer, but should not modify it in any way. The buffer
1741 will contain undecoded text of parts of the file. Each function
1742 should take one argument, SIZE, which says how many characters
1743 \(starting from point) it should look at.
1745 If one of these functions succeeds in determining a coding
1746 system, it should return that coding system. Otherwise, it
1747 should return nil.
1749 If a file has a `coding:' tag, that takes precedence over these
1750 functions, so they won't be called at all."
1751 :group 'files
1752 :group 'mule
1753 :type '(repeat function))
1755 (defvar set-auto-coding-for-load nil
1756 "Non-nil means look for `load-coding' property instead of `coding'.
1757 This is used for loading and byte-compiling Emacs Lisp files.")
1759 (defun auto-coding-alist-lookup (filename)
1760 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1761 (let ((alist auto-coding-alist)
1762 (case-fold-search (memq system-type '(windows-nt ms-dos cygwin)))
1763 coding-system)
1764 (while (and alist (not coding-system))
1765 (if (string-match (car (car alist)) filename)
1766 (setq coding-system (cdr (car alist)))
1767 (setq alist (cdr alist))))
1768 coding-system))
1770 (put 'enable-character-translation 'permanent-local t)
1771 (put 'enable-character-translation 'safe-local-variable 'booleanp)
1773 (defun find-auto-coding (filename size)
1774 "Find a coding system for a file FILENAME of which SIZE bytes follow point.
1775 These bytes should include at least the first 1k of the file
1776 and the last 3k of the file, but the middle may be omitted.
1778 The function checks FILENAME against the variable `auto-coding-alist'.
1779 If FILENAME doesn't match any entries in the variable, it checks the
1780 contents of the current buffer following point against
1781 `auto-coding-regexp-alist'. If no match is found, it checks for a
1782 `coding:' tag in the first one or two lines following point. If no
1783 `coding:' tag is found, it checks any local variables list in the last
1784 3K bytes out of the SIZE bytes. Finally, if none of these methods
1785 succeed, it checks to see if any function in `auto-coding-functions'
1786 gives a match.
1788 If a coding system is specifed, the return value is a cons
1789 \(CODING . SOURCE), where CODING is the specified coding system and
1790 SOURCE is a symbol `auto-coding-alist', `auto-coding-regexp-alist',
1791 `:coding', or `auto-coding-functions' indicating by what CODING is
1792 specified. Note that the validity of CODING is not checked;
1793 it's the caller's responsibility to check it.
1795 If nothing is specified, the return value is nil."
1796 (or (let ((coding-system (auto-coding-alist-lookup filename)))
1797 (if coding-system
1798 (cons coding-system 'auto-coding-alist)))
1799 ;; Try using `auto-coding-regexp-alist'.
1800 (let ((coding-system (auto-coding-regexp-alist-lookup (point)
1801 (+ (point) size))))
1802 (if coding-system
1803 (cons coding-system 'auto-coding-regexp-alist)))
1804 (let* ((case-fold-search t)
1805 (head-start (point))
1806 (head-end (+ head-start (min size 1024)))
1807 (tail-start (+ head-start (max (- size 3072) 0)))
1808 (tail-end (+ head-start size))
1809 coding-system head-found tail-found pos char-trans)
1810 ;; Try a short cut by searching for the string "coding:"
1811 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1812 (setq head-found (or (search-forward "coding:" head-end t)
1813 (search-forward "unibyte:" head-end t)
1814 (search-forward "enable-character-translation:"
1815 head-end t)))
1816 (if (and head-found (> head-found tail-start))
1817 ;; Head and tail are overlapped.
1818 (setq tail-found head-found)
1819 (goto-char tail-start)
1820 (setq tail-found (or (search-forward "coding:" tail-end t)
1821 (search-forward "unibyte:" tail-end t)
1822 (search-forward "enable-character-translation:"
1823 tail-end t))))
1825 ;; At first check the head.
1826 (when head-found
1827 (goto-char head-start)
1828 (setq head-end (set-auto-mode-1))
1829 (setq head-start (point))
1830 (when (and head-end (< head-found head-end))
1831 (goto-char head-start)
1832 (when (and set-auto-coding-for-load
1833 (re-search-forward
1834 "\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1835 head-end t))
1836 (setq coding-system 'raw-text))
1837 (when (and (not coding-system)
1838 (re-search-forward
1839 "\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1840 head-end t))
1841 (setq coding-system (intern (match-string 2))))
1842 (when (re-search-forward
1843 "\\(.*;\\)?[ \t]*enable-character-translation:[ \t]*\\([^ ;]+\\)"
1844 head-end t)
1845 (setq char-trans (match-string 2)))))
1847 ;; If no coding: tag in the head, check the tail.
1848 ;; Here we must pay attention to the case that the end-of-line
1849 ;; is just "\r" and we can't use "^" nor "$" in regexp.
1850 (when (and tail-found (or (not coding-system) (not char-trans)))
1851 (goto-char tail-start)
1852 (re-search-forward "[\r\n]\^L" tail-end t)
1853 (if (re-search-forward
1854 "[\r\n]\\([^[\r\n]*\\)[ \t]*Local Variables:[ \t]*\\([^\r\n]*\\)[\r\n]"
1855 tail-end t)
1856 ;; The prefix is what comes before "local variables:" in its
1857 ;; line. The suffix is what comes after "local variables:"
1858 ;; in its line.
1859 (let* ((prefix (regexp-quote (match-string 1)))
1860 (suffix (regexp-quote (match-string 2)))
1861 (re-coding
1862 (concat
1863 "[\r\n]" prefix
1864 ;; N.B. without the \n below, the regexp can
1865 ;; eat newlines.
1866 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1867 suffix "[\r\n]"))
1868 (re-unibyte
1869 (concat
1870 "[\r\n]" prefix
1871 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1872 suffix "[\r\n]"))
1873 (re-char-trans
1874 (concat
1875 "[\r\n]" prefix
1876 "[ \t]*enable-character-translation[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1877 suffix "[\r\n]"))
1878 (re-end
1879 (concat "[\r\n]" prefix "[ \t]*End *:[ \t]*" suffix
1880 "[\r\n]?"))
1881 (pos (1- (point))))
1882 (forward-char -1) ; skip back \r or \n.
1883 (re-search-forward re-end tail-end 'move)
1884 (setq tail-end (point))
1885 (goto-char pos)
1886 (when (and set-auto-coding-for-load
1887 (re-search-forward re-unibyte tail-end t))
1888 (setq coding-system 'raw-text))
1889 (when (and (not coding-system)
1890 (re-search-forward re-coding tail-end t))
1891 (setq coding-system (intern (match-string 1))))
1892 (when (and (not char-trans)
1893 (re-search-forward re-char-trans tail-end t))
1894 (setq char-trans (match-string 1))))))
1895 (if coding-system
1896 ;; If the coding-system name ends with "!", remove it and
1897 ;; set char-trans to "nil".
1898 (let ((name (symbol-name coding-system)))
1899 (if (= (aref name (1- (length name))) ?!)
1900 (setq coding-system (intern (substring name 0 -1))
1901 char-trans "nil"))))
1902 (when (and char-trans
1903 (not (setq char-trans (intern char-trans))))
1904 (make-local-variable 'enable-character-translation)
1905 (setq enable-character-translation nil))
1906 (if coding-system
1907 (cons coding-system :coding)))
1908 ;; Finally, try all the `auto-coding-functions'.
1909 (let ((funcs auto-coding-functions)
1910 (coding-system nil))
1911 (while (and funcs (not coding-system))
1912 (setq coding-system (condition-case e
1913 (save-excursion
1914 (goto-char (point-min))
1915 (funcall (pop funcs) size))
1916 (error nil))))
1917 (if coding-system
1918 (cons coding-system 'auto-coding-functions)))))
1920 (defun set-auto-coding (filename size)
1921 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1922 See `find-auto-coding' for how the coding system is found.
1923 Return nil if an invalid coding system is found.
1925 The variable `set-auto-coding-function' (which see) is set to this
1926 function by default."
1927 (let ((found (find-auto-coding filename size)))
1928 (if (and found (coding-system-p (car found)))
1929 (car found))))
1931 (setq set-auto-coding-function 'set-auto-coding)
1933 (defun after-insert-file-set-coding (inserted &optional visit)
1934 "Set `buffer-file-coding-system' of current buffer after text is inserted.
1935 INSERTED is the number of characters that were inserted, as figured
1936 in the situation before this function. Return the number of characters
1937 inserted, as figured in the situation after. The two numbers can be
1938 different if the buffer has become unibyte.
1939 The optional second arg VISIT non-nil means that we are visiting a file."
1940 (if (and visit
1941 coding-system-for-read
1942 (not (eq coding-system-for-read 'auto-save-coding)))
1943 (setq buffer-file-coding-system-explicit
1944 (cons coding-system-for-read nil)))
1945 (if last-coding-system-used
1946 (let ((coding-system
1947 (find-new-buffer-file-coding-system last-coding-system-used)))
1948 (if coding-system
1949 (setq buffer-file-coding-system coding-system))))
1950 inserted)
1952 ;; The coding-spec and eol-type of coding-system returned is decided
1953 ;; independently in the following order.
1954 ;; 1. That of buffer-file-coding-system locally bound.
1955 ;; 2. That of CODING.
1957 (defun find-new-buffer-file-coding-system (coding)
1958 "Return a coding system for a buffer when a file of CODING is inserted.
1959 The local variable `buffer-file-coding-system' of the current buffer
1960 is set to the returned value.
1961 Return nil if there's no need to set `buffer-file-coding-system'."
1962 (let (local-coding local-eol
1963 found-coding found-eol
1964 new-coding new-eol)
1965 (if (null coding)
1966 ;; Nothing found about coding.
1969 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1970 ;; and LOCAL-CODING.
1971 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1972 (if (null (numberp local-eol))
1973 ;; But eol-type is not yet set.
1974 (setq local-eol nil))
1975 (if (and buffer-file-coding-system
1976 (not (eq (coding-system-type buffer-file-coding-system)
1977 'undecided)))
1978 (setq local-coding (coding-system-base buffer-file-coding-system)))
1980 (if (and (local-variable-p 'buffer-file-coding-system)
1981 local-eol local-coding)
1982 ;; The current buffer has already set full coding-system, we
1983 ;; had better not change it.
1986 (setq found-eol (coding-system-eol-type coding))
1987 (if (null (numberp found-eol))
1988 ;; But eol-type is not found.
1989 ;; If EOL conversions are inhibited, force unix eol-type.
1990 (setq found-eol (if inhibit-eol-conversion 0)))
1991 (setq found-coding (coding-system-base coding))
1993 (if (and (not found-eol) (eq found-coding 'undecided))
1994 ;; No valid coding information found.
1997 ;; Some coding information (eol or text) found.
1999 ;; The local setting takes precedence over the found one.
2000 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
2001 (or local-coding found-coding)
2002 (or found-coding local-coding)))
2003 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
2004 (or local-eol found-eol)
2005 (or found-eol local-eol)))
2007 (let ((eol-type (coding-system-eol-type new-coding)))
2008 (if (and (numberp new-eol) (vectorp eol-type))
2009 (aref eol-type new-eol)
2010 new-coding)))))))
2012 (defun modify-coding-system-alist (target-type regexp coding-system)
2013 "Modify one of look up tables for finding a coding system on I/O operation.
2014 There are three of such tables, `file-coding-system-alist',
2015 `process-coding-system-alist', and `network-coding-system-alist'.
2017 TARGET-TYPE specifies which of them to modify.
2018 If it is `file', it affects `file-coding-system-alist' (which see).
2019 If it is `process', it affects `process-coding-system-alist' (which see).
2020 If it is `network', it affects `network-coding-system-alist' (which see).
2022 REGEXP is a regular expression matching a target of I/O operation.
2023 The target is a file name if TARGET-TYPE is `file', a program name if
2024 TARGET-TYPE is `process', or a network service name or a port number
2025 to connect to if TARGET-TYPE is `network'.
2027 CODING-SYSTEM is a coding system to perform code conversion on the I/O
2028 operation, or a cons cell (DECODING . ENCODING) specifying the coding
2029 systems for decoding and encoding respectively, or a function symbol
2030 which, when called, returns such a cons cell."
2031 (or (memq target-type '(file process network))
2032 (error "Invalid target type: %s" target-type))
2033 (or (stringp regexp)
2034 (and (eq target-type 'network) (integerp regexp))
2035 (error "Invalid regular expression: %s" regexp))
2036 (if (symbolp coding-system)
2037 (if (not (fboundp coding-system))
2038 (progn
2039 (check-coding-system coding-system)
2040 (setq coding-system (cons coding-system coding-system))))
2041 (check-coding-system (car coding-system))
2042 (check-coding-system (cdr coding-system)))
2043 (cond ((eq target-type 'file)
2044 (let ((slot (assoc regexp file-coding-system-alist)))
2045 (if slot
2046 (setcdr slot coding-system)
2047 (setq file-coding-system-alist
2048 (cons (cons regexp coding-system)
2049 file-coding-system-alist)))))
2050 ((eq target-type 'process)
2051 (let ((slot (assoc regexp process-coding-system-alist)))
2052 (if slot
2053 (setcdr slot coding-system)
2054 (setq process-coding-system-alist
2055 (cons (cons regexp coding-system)
2056 process-coding-system-alist)))))
2058 (let ((slot (assoc regexp network-coding-system-alist)))
2059 (if slot
2060 (setcdr slot coding-system)
2061 (setq network-coding-system-alist
2062 (cons (cons regexp coding-system)
2063 network-coding-system-alist)))))))
2065 (defun decode-coding-inserted-region (from to filename
2066 &optional visit beg end replace)
2067 "Decode the region between FROM and TO as if it is read from file FILENAME.
2068 The idea is that the text between FROM and TO was just inserted somehow.
2069 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
2070 of the function `insert-file-contents'.
2071 Part of the job of this function is setting `buffer-undo-list' appropriately."
2072 (save-excursion
2073 (save-restriction
2074 (let ((coding coding-system-for-read)
2075 undo-list-saved)
2076 (if visit
2077 ;; Temporarily turn off undo recording, if we're decoding the
2078 ;; text of a visited file.
2079 (setq buffer-undo-list t)
2080 ;; Otherwise, if we can recognize the undo elt for the insertion,
2081 ;; remove it and get ready to replace it later.
2082 ;; In the mean time, turn off undo recording.
2083 (let ((last (car-safe buffer-undo-list)))
2084 (if (and (consp last) (eql (car last) from) (eql (cdr last) to))
2085 (setq undo-list-saved (cdr buffer-undo-list)
2086 buffer-undo-list t))))
2087 (narrow-to-region from to)
2088 (goto-char (point-min))
2089 (or coding
2090 (setq coding (funcall set-auto-coding-function
2091 filename (- (point-max) (point-min)))))
2092 (or coding
2093 (setq coding (car (find-operation-coding-system
2094 'insert-file-contents
2095 (cons filename (current-buffer))
2096 visit beg end replace))))
2097 (if (coding-system-p coding)
2098 (or enable-multibyte-characters
2099 (setq coding
2100 (coding-system-change-text-conversion coding 'raw-text)))
2101 (setq coding nil))
2102 (if coding
2103 (decode-coding-region (point-min) (point-max) coding)
2104 (setq last-coding-system-used coding))
2105 ;; If we're decoding the text of a visited file,
2106 ;; the undo list should start out empty.
2107 (if visit
2108 (setq buffer-undo-list nil)
2109 ;; If we decided to replace the undo entry for the insertion,
2110 ;; do so now.
2111 (if undo-list-saved
2112 (setq buffer-undo-list
2113 (cons (cons from (point-max)) undo-list-saved))))))))
2115 (defun recode-region (start end new-coding coding)
2116 "Re-decode the region (previously decoded by CODING) by NEW-CODING."
2117 (interactive
2118 (list (region-beginning) (region-end)
2119 (read-coding-system "Text was really in: ")
2120 (let ((coding (or buffer-file-coding-system last-coding-system-used)))
2121 (read-coding-system
2122 (concat "But was interpreted as"
2123 (if coding (format " (default %S): " coding) ": "))
2124 coding))))
2125 (or (and new-coding coding)
2126 (error "Coding system not specified"))
2127 ;; Check it before we encode the region.
2128 (check-coding-system new-coding)
2129 (save-restriction
2130 (narrow-to-region start end)
2131 (encode-coding-region (point-min) (point-max) coding)
2132 (decode-coding-region (point-min) (point-max) new-coding))
2133 (if (region-active-p)
2134 (deactivate-mark)))
2136 (defun make-translation-table (&rest args)
2137 "Make a translation table from arguments.
2138 A translation table is a char table intended for character
2139 translation in CCL programs.
2141 Each argument is a list of elements of the form (FROM . TO), where FROM
2142 is a character to be translated to TO.
2144 The arguments and forms in each argument are processed in the given
2145 order, and if a previous form already translates TO to some other
2146 character, say TO-ALT, FROM is also translated to TO-ALT."
2147 (let ((table (make-char-table 'translation-table))
2148 revlist)
2149 (dolist (elts args)
2150 (dolist (elt elts)
2151 (let ((from (car elt))
2152 (to (cdr elt))
2153 to-alt rev-from rev-to)
2154 ;; If we have already translated TO to TO-ALT, FROM should
2155 ;; also be translated to TO-ALT.
2156 (if (setq to-alt (aref table to))
2157 (setq to to-alt))
2158 (aset table from to)
2159 ;; If we have already translated some chars to FROM, they
2160 ;; should also be translated to TO.
2161 (when (setq rev-from (assq from revlist))
2162 (dolist (elt (cdr rev-from))
2163 (aset table elt to))
2164 (setq revlist (delq rev-from revlist)
2165 rev-from (cdr rev-from)))
2166 ;; Now update REVLIST.
2167 (setq rev-to (assq to revlist))
2168 (if rev-to
2169 (setcdr rev-to (cons from (cdr rev-to)))
2170 (setq rev-to (list to from)
2171 revlist (cons rev-to revlist)))
2172 (if rev-from
2173 (setcdr rev-to (append rev-from (cdr rev-to)))))))
2174 ;; Return TABLE just created.
2175 (set-char-table-extra-slot table 1 1)
2176 table))
2178 (defun make-translation-table-from-vector (vec)
2179 "Make translation table from decoding vector VEC.
2180 VEC is an array of 256 elements to map unibyte codes to multibyte
2181 characters. Elements may be nil for undefined code points."
2182 (let ((table (make-char-table 'translation-table))
2183 (rev-table (make-char-table 'translation-table))
2185 (dotimes (i 256)
2186 (setq ch (aref vec i))
2187 (when ch
2188 (aset table i ch)
2189 (if (>= ch 256)
2190 (aset rev-table ch i))))
2191 (set-char-table-extra-slot table 0 rev-table)
2192 (set-char-table-extra-slot table 1 1)
2193 (set-char-table-extra-slot rev-table 1 1)
2194 table))
2196 (defun make-translation-table-from-alist (alist)
2197 "Make translation table from N<->M mapping in ALIST.
2198 ALIST is an alist, each element has the form (FROM . TO).
2199 FROM and TO are a character or a vector of characters.
2200 If FROM is a character, that character is translated to TO.
2201 If FROM is a vector of characters, that sequence is translated to TO.
2202 The first extra-slot of the value is a translation table for reverse mapping."
2203 (let ((tables (vector (make-char-table 'translation-table)
2204 (make-char-table 'translation-table)))
2205 table max-lookup from to idx val)
2206 (dotimes (i 2)
2207 (setq table (aref tables i))
2208 (setq max-lookup 1)
2209 (dolist (elt alist)
2210 (if (= i 0)
2211 (setq from (car elt) to (cdr elt))
2212 (setq from (cdr elt) to (car elt)))
2213 (if (characterp from)
2214 (setq idx from)
2215 (setq idx (aref from 0)
2216 max-lookup (max max-lookup (length from))))
2217 (setq val (aref table idx))
2218 (if val
2219 (progn
2220 (or (consp val)
2221 (setq val (list (cons (vector idx) val))))
2222 (if (characterp from)
2223 (setq from (vector from)))
2224 (setq val (nconc val (list (cons from to)))))
2225 (if (characterp from)
2226 (setq val to)
2227 (setq val (list (cons from to)))))
2228 (aset table idx val))
2229 (set-char-table-extra-slot table 1 max-lookup))
2230 (set-char-table-extra-slot (aref tables 0) 0 (aref tables 1))
2231 (aref tables 0)))
2233 (defun define-translation-table (symbol &rest args)
2234 "Define SYMBOL as the name of translation table made by ARGS.
2235 This sets up information so that the table can be used for
2236 translations in a CCL program.
2238 If the first element of ARGS is a char-table whose purpose is
2239 `translation-table', just define SYMBOL to name it. (Note that this
2240 function does not bind SYMBOL.)
2242 Any other ARGS should be suitable as arguments of the function
2243 `make-translation-table' (which see).
2245 This function sets properties `translation-table' and
2246 `translation-table-id' of SYMBOL to the created table itself and the
2247 identification number of the table respectively. It also registers
2248 the table in `translation-table-vector'."
2249 (let ((table (if (and (char-table-p (car args))
2250 (eq (char-table-subtype (car args))
2251 'translation-table))
2252 (car args)
2253 (apply 'make-translation-table args)))
2254 (len (length translation-table-vector))
2255 (id 0)
2256 (done nil))
2257 (put symbol 'translation-table table)
2258 (while (not done)
2259 (if (>= id len)
2260 (setq translation-table-vector
2261 (vconcat translation-table-vector (make-vector len nil))))
2262 (let ((slot (aref translation-table-vector id)))
2263 (if (or (not slot)
2264 (eq (car slot) symbol))
2265 (progn
2266 (aset translation-table-vector id (cons symbol table))
2267 (setq done t))
2268 (setq id (1+ id)))))
2269 (put symbol 'translation-table-id id)
2270 id))
2272 (defun translate-region (start end table)
2273 "From START to END, translate characters according to TABLE.
2274 TABLE is a string or a char-table.
2275 If TABLE is a string, the Nth character in it is the mapping
2276 for the character with code N.
2277 If TABLE is a char-table, the element for character N is the mapping
2278 for the character with code N.
2279 It returns the number of characters changed."
2280 (interactive
2281 (list (region-beginning)
2282 (region-end)
2283 (let (table l)
2284 (dotimes (i (length translation-table-vector))
2285 (if (consp (aref translation-table-vector i))
2286 (push (list (symbol-name
2287 (car (aref translation-table-vector i)))) l)))
2288 (if (not l)
2289 (error "No translation table defined"))
2290 (while (not table)
2291 (setq table (completing-read "Translation table: " l nil t)))
2292 (intern table))))
2293 (if (symbolp table)
2294 (let ((val (get table 'translation-table)))
2295 (or (char-table-p val)
2296 (error "Invalid translation table name: %s" table))
2297 (setq table val)))
2298 (translate-region-internal start end table))
2300 (defmacro with-category-table (table &rest body)
2301 "Execute BODY like `progn' with TABLE the current category table.
2302 The category table of the current buffer is saved, BODY is evaluated,
2303 then the saved table is restored, even in case of an abnormal exit.
2304 Value is what BODY returns."
2305 (declare (indent 1) (debug t))
2306 (let ((old-table (make-symbol "old-table"))
2307 (old-buffer (make-symbol "old-buffer")))
2308 `(let ((,old-table (category-table))
2309 (,old-buffer (current-buffer)))
2310 (unwind-protect
2311 (progn
2312 (set-category-table ,table)
2313 ,@body)
2314 (with-current-buffer ,old-buffer
2315 (set-category-table ,old-table))))))
2317 (defun define-translation-hash-table (symbol table)
2318 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2320 Analogous to `define-translation-table', but updates
2321 `translation-hash-table-vector' and the table is for use in the CCL
2322 `lookup-integer' and `lookup-character' functions."
2323 (unless (and (symbolp symbol)
2324 (hash-table-p table))
2325 (error "Bad args to define-translation-hash-table"))
2326 (let ((len (length translation-hash-table-vector))
2327 (id 0)
2328 done)
2329 (put symbol 'translation-hash-table table)
2330 (while (not done)
2331 (if (>= id len)
2332 (setq translation-hash-table-vector
2333 (vconcat translation-hash-table-vector [nil])))
2334 (let ((slot (aref translation-hash-table-vector id)))
2335 (if (or (not slot)
2336 (eq (car slot) symbol))
2337 (progn
2338 (aset translation-hash-table-vector id (cons symbol table))
2339 (setq done t))
2340 (setq id (1+ id)))))
2341 (put symbol 'translation-hash-table-id id)
2342 id))
2344 ;;; Initialize some variables.
2346 (put 'use-default-ascent 'char-table-extra-slots 0)
2347 (setq use-default-ascent (make-char-table 'use-default-ascent))
2348 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2349 (setq ignore-relative-composition
2350 (make-char-table 'ignore-relative-composition))
2352 (make-obsolete 'set-char-table-default
2353 "generic characters no longer exist." "23.1")
2355 ;;; Built-in auto-coding-functions:
2357 (defun sgml-xml-auto-coding-function (size)
2358 "Determine whether the buffer is XML, and if so, its encoding.
2359 This function is intended to be added to `auto-coding-functions'."
2360 (setq size (+ (point) size))
2361 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" size t)
2362 (let ((end (save-excursion
2363 ;; This is a hack.
2364 (re-search-forward "[\"']\\s-*\\?>" size t))))
2365 (when end
2366 (if (re-search-forward "encoding=[\"']\\(.+?\\)[\"']" end t)
2367 (let* ((match (match-string 1))
2368 (sym (intern (downcase match))))
2369 (if (coding-system-p sym)
2371 (message "Warning: unknown coding system \"%s\"" match)
2372 nil))
2373 ;; Files without an encoding tag should be UTF-8. But users
2374 ;; may be naive about encodings, and have saved the file from
2375 ;; another editor that does not help them get the encoding right.
2376 ;; Detect the encoding and warn the user if it is detected as
2377 ;; something other than UTF-8.
2378 (let ((detected
2379 (with-coding-priority '(utf-8)
2380 (coding-system-base
2381 (detect-coding-region (point-min) size t)))))
2382 ;; Pure ASCII always comes back as undecided.
2383 (if (memq detected '(utf-8 undecided))
2384 'utf-8
2385 (warn "File contents detected as %s.
2386 Consider adding an encoding attribute to the xml declaration,
2387 or saving as utf-8, as mandated by the xml specification." detected)
2388 detected)))))))
2390 (defun sgml-html-meta-auto-coding-function (size)
2391 "If the buffer has an HTML meta tag, use it to determine encoding.
2392 This function is intended to be added to `auto-coding-functions'."
2393 (let ((case-fold-search t))
2394 (setq size (min (+ (point) size)
2395 (save-excursion
2396 ;; Limit the search by the end of the HTML header.
2397 (or (search-forward "</head>" (+ (point) size) t)
2398 ;; In case of no header, search only 10 lines.
2399 (forward-line 10))
2400 (point))))
2401 ;; Make sure that the buffer really contains an HTML document, by
2402 ;; checking that it starts with a doctype or a <HTML> start tag
2403 ;; (allowing for whitespace at bob). Note: 'DOCTYPE NETSCAPE' is
2404 ;; useful for Mozilla bookmark files.
2405 (when (and (re-search-forward "\\`[[:space:]\n]*\\(<!doctype[[:space:]\n]+\\(html\\|netscape\\)\\|<html\\)" size t)
2406 (re-search-forward "<meta\\s-+\\(http-equiv=[\"']?content-type[\"']?\\s-+content=[\"']text/\\sw+;\\s-*\\)?charset=[\"']?\\(.+?\\)[\"'\\s-/>]" size t))
2407 (let* ((match (match-string 2))
2408 (sym (intern (downcase match))))
2409 (if (coding-system-p sym)
2411 (message "Warning: unknown coding system \"%s\"" match)
2412 nil)))))
2414 (defun xml-find-file-coding-system (args)
2415 "Determine the coding system of an XML file without a declaration.
2416 Strictly speaking, the file should be utf-8, but mistakes are
2417 made, and there are genuine cases where XML fragments are saved,
2418 with the encoding properly specified in a master document, or
2419 added by processing software."
2420 (if (eq (car args) 'insert-file-contents)
2421 (let ((detected
2422 (with-coding-priority '(utf-8)
2423 (coding-system-base
2424 (detect-coding-region (point-min) (point-max) t)))))
2425 ;; Pure ASCII always comes back as undecided.
2426 (cond
2427 ((memq detected '(utf-8 undecided))
2428 'utf-8)
2429 ((eq detected 'utf-16le-with-signature) 'utf-16le-with-signature)
2430 ((eq detected 'utf-16be-with-signature) 'utf-16be-with-signature)
2432 (warn "File contents detected as %s.
2433 Consider adding an xml declaration with the encoding specified,
2434 or saving as utf-8, as mandated by the xml specification." detected)
2435 detected)))
2436 ;; Don't interfere with the user's wishes for saving the buffer.
2437 ;; We did what we could when the buffer was created to ensure the
2438 ;; correct encoding was used, or the user was warned, so any
2439 ;; non-conformity here is deliberate on the part of the user.
2440 'undecided))
2443 (provide 'mule)
2445 ;;; mule.el ends here