1 ;;; mule.el --- basic commands for multilingual environment
3 ;; Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006,
4 ;; 2007, 2008, 2009, 2010
5 ;; Free Software Foundation, Inc.
6 ;; Copyright (C) 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003, 2004,
7 ;; 2005, 2006, 2007, 2008, 2009, 2010
8 ;; National Institute of Advanced Industrial Science and Technology (AIST)
9 ;; Registration Number H14PRO021
11 ;; National Institute of Advanced Industrial Science and Technology (AIST)
12 ;; Registration Number H13PRO009
14 ;; Keywords: mule, multilingual, character set, coding system
16 ;; This file is part of GNU Emacs.
18 ;; GNU Emacs is free software: you can redistribute it and/or modify
19 ;; it under the terms of the GNU General Public License as published by
20 ;; the Free Software Foundation, either version 3 of the License, or
21 ;; (at your option) any later version.
23 ;; GNU Emacs is distributed in the hope that it will be useful,
24 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
25 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
26 ;; GNU General Public License for more details.
28 ;; You should have received a copy of the GNU General Public License
29 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
35 (defconst mule-version "6.0 (HANACHIRUSATO)" "\
36 Version number and name of this version of MULE (multilingual environment).")
38 (defconst mule-version-date "2003.9.1" "\
39 Distribution date of this version of MULE (multilingual environment).")
44 ;; Backward compatibility code for handling emacs-mule charsets.
45 (defvar private-char-area-1-min #xF0000)
46 (defvar private-char-area-1-max #xFFFFE)
47 (defvar private-char-area-2-min #x100000)
48 (defvar private-char-area-2-max #x10FFFE)
50 ;; Table of emacs-mule charsets indexed by their emacs-mule ID.
51 (defvar emacs-mule-charset-table (make-vector 256 nil))
52 (aset emacs-mule-charset-table 0 'ascii)
54 ;; Convert the argument of old-style call of define-charset to a
55 ;; property list used by the new-style.
56 ;; INFO-VECTOR is a vector of the format:
57 ;; [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE
58 ;; SHORT-NAME LONG-NAME DESCRIPTION]
60 (defun convert-define-charset-argument (emacs-mule-id info-vector)
61 (let* ((dim (aref info-vector 0))
62 (chars (aref info-vector 1))
63 (total (if (= dim 1) chars (* chars chars)))
64 (code-space (if (= dim 1) (if (= chars 96) [32 127] [33 126])
65 (if (= chars 96) [32 127 32 127] [33 126 33 126])))
67 (if (integerp emacs-mule-id)
68 (or (= emacs-mule-id 0)
69 (and (>= emacs-mule-id 129) (< emacs-mule-id 256))
70 (error "Invalid CHARSET-ID: %d" emacs-mule-id))
72 (if (= dim 1) (setq from-id 160 to-id 224)
73 (setq from-id 224 to-id 255))
74 (while (and (< from-id to-id)
75 (not (aref emacs-mule-charset-table from-id)))
76 (setq from-id (1+ from-id)))
78 (error "No more room for the new Emacs-mule charset"))
79 (setq emacs-mule-id from-id)))
80 (if (> (- private-char-area-1-max private-char-area-1-min) total)
81 (setq code-offset private-char-area-1-min
82 private-char-area-1-min (+ private-char-area-1-min total))
83 (if (> (- private-char-area-2-max private-char-area-2-min) total)
84 (setq code-offset private-char-area-2-min
85 private-char-area-2-min (+ private-char-area-2-min total))
86 (error "No more space for a new charset")))
88 :code-space code-space
89 :iso-final-char (aref info-vector 4)
90 :code-offset code-offset
91 :emacs-mule-id emacs-mule-id)))
93 (defun define-charset (name docstring &rest props)
94 "Define NAME (symbol) as a charset with DOCSTRING.
95 The remaining arguments must come in pairs ATTRIBUTE VALUE. ATTRIBUTE
96 may be any symbol. The following have special meanings, and one of
97 `:code-offset', `:map', `:subset', `:superset' must be specified.
101 VALUE must be a short string to identify the charset. If omitted,
106 VALUE must be a string longer than `:short-name' to identify the
107 charset. If omitted, the value of the `:short-name' attribute is used.
111 VALUE must be an integer 0, 1, 2, or 3, specifying the dimension of
112 code-points of the charsets. If omitted, it is calculated from the
113 value of the `:code-space' attribute.
117 VALUE must be a vector of length at most 8 specifying the byte code
118 range of each dimension in this format:
119 [ MIN-1 MAX-1 MIN-2 MAX-2 ... ]
120 where MIN-N is the minimum byte value of Nth dimension of code-point,
121 MAX-N is the maximum byte value of that.
125 VALUE must be an integer specifying the mininum code point of the
126 charset. If omitted, it is calculated from `:code-space'. VALUE may
127 be a cons (HIGH . LOW), where HIGH is the most significant 16 bits of
128 the code point and LOW is the least significant 16 bits.
132 VALUE must be an integer specifying the maxinum code point of the
133 charset. If omitted, it is calculated from `:code-space'. VALUE may
134 be a cons (HIGH . LOW), where HIGH is the most significant 16 bits of
135 the code point and LOW is the least significant 16 bits.
139 VALUE must be a character in the range 32 to 127 (inclusive)
140 specifying the final char of the charset for ISO-2022 encoding. If
141 omitted, the charset can't be encoded by ISO-2022 based
144 `:iso-revision-number'
146 VALUE must be an integer in the range 0..63, specifying the revision
147 number of the charset for ISO-2022 encoding.
151 VALUE must be an integer of 0, 129..255. If omitted, the charset
152 can't be encoded by coding-systems of type `emacs-mule'.
154 `:ascii-compatible-p'
156 VALUE must be nil or t (default nil). If VALUE is t, the charset is
157 compatible with ASCII, i.e. the first 128 code points map to ASCII.
161 VALUE must be nil or t. If the VALUE is t, the charset is
162 supplementary, which means it is used only as a parent or a
163 subset of some other charset, or it is provided just for backward
168 VALUE must be a nonnegative integer that can be used as an invalid
169 code point of the charset. If the minimum code is 0 and the maximum
170 code is greater than Emacs' maximum integer value, `:invalid-code'
171 should not be omitted.
175 VALUE must be an integer added to the index number of a character to
176 get the corresponding character code.
180 VALUE must be vector or string.
182 If it is a vector, the format is [ CODE-1 CHAR-1 CODE-2 CHAR-2 ... ],
183 where CODE-n is a code-point of the charset, and CHAR-n is the
184 corresponding character code.
186 If it is a string, it is a name of file that contains the above
187 information. Each line of the file must be this format:
189 where XXX is a hexadecimal representation of CODE-n and YYY is a
190 hexadecimal representation of CHAR-n. A line starting with `#' is a
195 VALUE must be a list:
196 ( PARENT MIN-CODE MAX-CODE OFFSET )
197 PARENT is a parent charset. MIN-CODE and MAX-CODE specify the range
198 of characters inherited from the parent. OFFSET is an integer value
199 to add to a code point of the parent charset to get the corresponding
200 code point of this charset.
204 VALUE must be a list of parent charsets. The charset inherits
205 characters from them. Each element of the list may be a cons (PARENT
206 . OFFSET), where PARENT is a parent charset, and OFFSET is an offset
207 value to add to a code point of PARENT to get the corresponding code
208 point of this charset.
212 VALUE must be vector or string.
214 If it is a vector, the format is [ CODE-1 CHAR-1 CODE-2 CHAR-2 ... ],
215 where CODE-n is a code-point of the charset, and CHAR-n is the
216 corresponding Unicode character code.
218 If it is a string, it is a name of file that contains the above
219 information. The file format is the same as what described for `:map'
221 (when (vectorp (car props))
223 ;; (define-charset CHARSET-ID CHARSET-SYMBOL INFO-VECTOR)
224 ;; Convert the argument to make it fit with the current style.
225 (let ((vec (car props)))
226 (setq props (convert-define-charset-argument name vec)
228 docstring (aref vec 8))))
229 (let ((attrs (mapcar 'list '(:dimension
246 ;; If :dimension is omitted, get the dimension from :code-space.
247 (let ((dimension (plist-get props :dimension)))
249 (let ((code-space (plist-get props :code-space)))
250 (setq dimension (if code-space (/ (length code-space) 2) 4))
251 (setq props (plist-put props :dimension dimension)))))
253 (let ((code-space (plist-get props :code-space)))
255 (let ((dimension (plist-get props :dimension)))
256 (setq code-space (make-vector 8 0))
257 (dotimes (i dimension)
258 (aset code-space (1+ (* i 2)) #xFF))
259 (setq props (plist-put props :code-space code-space)))))
261 ;; If :emacs-mule-id is specified, update emacs-mule-charset-table.
262 (let ((emacs-mule-id (plist-get props :emacs-mule-id)))
263 (if (integerp emacs-mule-id)
264 (aset emacs-mule-charset-table emacs-mule-id name)))
267 (setcdr slot (purecopy (plist-get props (car slot)))))
269 ;; Make sure that the value of :code-space is a vector of 8
271 (let* ((slot (assq :code-space attrs))
276 (vconcat val (make-vector (- 8 len) 0)))))
278 ;; Add :name and :docstring properties to PROPS.
280 (cons :name (cons name (cons :docstring (cons (purecopy docstring) props)))))
281 (or (plist-get props :short-name)
282 (plist-put props :short-name (symbol-name name)))
283 (or (plist-get props :long-name)
284 (plist-put props :long-name (plist-get props :short-name)))
285 ;; We can probably get a worthwhile amount in purespace.
287 (mapcar (lambda (elt)
292 (setcdr (assq :plist attrs) props)
294 (apply 'define-charset-internal name (mapcar 'cdr attrs))))
297 (defun load-with-code-conversion (fullname file &optional noerror nomessage)
298 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
299 The file contents are decoded before evaluation if necessary.
300 If optional third arg NOERROR is non-nil,
301 report no error if FILE doesn't exist.
302 Print messages at start and end of loading unless
303 optional fourth arg NOMESSAGE is non-nil.
304 Return t if file exists."
305 (if (null (file-readable-p fullname))
307 (signal 'file-error (list "Cannot open load file" file)))
308 ;; Read file with code conversion, and then eval.
310 ;; We can't use `generate-new-buffer' because files.el
311 ;; is not yet loaded.
312 (get-buffer-create (generate-new-buffer-name " *load*")))
314 (source (save-match-data (string-match "\\.el\\'" fullname))))
317 (message "Loading %s (source)..." file)
318 (message "Loading %s..." file)))
320 (push (purecopy file) preloaded-file-list))
322 (let ((load-file-name fullname)
323 (set-auto-coding-for-load t)
324 (inhibit-file-name-operation nil))
325 (with-current-buffer buffer
326 ;; So that we don't get completely screwed if the
327 ;; file is encoded in some complicated character set,
328 ;; read it with real decoding, as a multibyte buffer,
329 ;; even if this is a --unibyte Emacs session.
330 (set-buffer-multibyte t)
331 ;; Don't let deactivate-mark remain set.
332 (let (deactivate-mark)
333 (insert-file-contents fullname))
334 ;; If the loaded file was inserted with no-conversion or
335 ;; raw-text coding system, make the buffer unibyte.
336 ;; Otherwise, eval-buffer might try to interpret random
337 ;; binary junk as multibyte characters.
338 (if (and enable-multibyte-characters
339 (or (eq (coding-system-type last-coding-system-used)
341 (set-buffer-multibyte nil))
342 ;; Make `kill-buffer' quiet.
343 (set-buffer-modified-p nil))
344 ;; Have the original buffer current while we eval.
345 (eval-buffer buffer nil
346 ;; This is compatible with what `load' does.
347 (if purify-flag file fullname)
348 ;; If this Emacs is running with --unibyte,
349 ;; convert multibyte strings to unibyte
350 ;; after reading them.
351 ;; (not (default-value 'enable-multibyte-characters))
354 (let (kill-buffer-hook kill-buffer-query-functions)
355 (kill-buffer buffer)))
356 (do-after-load-evaluation fullname)
358 (unless (or nomessage noninteractive)
360 (message "Loading %s (source)...done" file)
361 (message "Loading %s...done" file)))
364 (defun charset-info (charset)
365 "Return a vector of information of CHARSET.
366 This function is provided for backward compatibility.
368 The elements of the vector are:
369 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
370 LEADING-CODE-BASE, LEADING-CODE-EXT,
371 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
372 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
375 CHARSET-ID is always 0.
377 DIMENSION is the number of bytes of a code-point of the charset:
379 CHARS is the number of characters in a dimension:
382 DIRECTION is always 0.
383 LEADING-CODE-BASE is always 0.
384 LEADING-CODE-EXT is always 0.
385 ISO-FINAL-CHAR (character) is the final character of the
386 corresponding ISO 2022 charset. If the charset is not assigned
387 any final character, the value is -1.
388 ISO-GRAPHIC-PLANE is always 0.
389 REVERSE-CHARSET is always -1.
390 SHORT-NAME (string) is the short name to refer to the charset.
391 LONG-NAME (string) is the long name to refer to the charset
392 DESCRIPTION (string) is the description string of the charset.
393 PLIST (property list) may contain any type of information a user
394 want to put and get by functions `put-charset-property' and
395 `get-charset-property' respectively."
398 (charset-dimension charset)
399 (charset-chars charset)
404 (charset-iso-final-char charset)
407 (get-charset-property charset :short-name)
408 (get-charset-property charset :short-name)
409 (charset-description charset)
410 (charset-plist charset)))
412 ;; It is better not to use backquote in this file,
413 ;; because that makes a bootstrapping problem
414 ;; if you need to recompile all the Lisp files using interpreted code.
416 (defun charset-id (charset)
417 "Always return 0. This is provided for backward compatibility."
419 (make-obsolete 'charset-id "do not use it." "23.1")
421 (defmacro charset-bytes (charset)
422 "Always return 0. This is provided for backward compatibility."
424 (make-obsolete 'charset-bytes "do not use it." "23.1")
426 (defun get-charset-property (charset propname)
427 "Return the value of CHARSET's PROPNAME property.
428 This is the last value stored with
429 (put-charset-property CHARSET PROPNAME VALUE)."
430 (plist-get (charset-plist charset) propname))
432 (defun put-charset-property (charset propname value)
433 "Set CHARSETS's PROPNAME property to value VALUE.
434 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
435 (set-charset-plist charset
436 (plist-put (charset-plist charset) propname
441 (defun charset-description (charset)
442 "Return description string of CHARSET."
443 (plist-get (charset-plist charset) :docstring))
445 (defun charset-dimension (charset)
446 "Return dimension of CHARSET."
447 (plist-get (charset-plist charset) :dimension))
449 (defun charset-chars (charset &optional dimension)
450 "Return number of characters contained in DIMENSION of CHARSET.
451 DIMENSION defaults to the first dimension."
452 (unless dimension (setq dimension 1))
453 (let ((code-space (plist-get (charset-plist charset) :code-space)))
454 (1+ (- (aref code-space (1- (* 2 dimension)))
455 (aref code-space (- (* 2 dimension) 2))))))
457 (defun charset-iso-final-char (charset)
458 "Return ISO-2022 final character of CHARSET.
459 Return -1 if charset isn't an ISO 2022 one."
460 (or (plist-get (charset-plist charset) :iso-final-char)
463 (defmacro charset-short-name (charset)
464 "Return short name of CHARSET."
465 (plist-get (charset-plist charset) :short-name))
467 (defmacro charset-long-name (charset)
468 "Return long name of CHARSET."
469 (plist-get (charset-plist charset) :long-name))
471 (defun charset-list ()
472 "Return list of all charsets ever defined."
474 (make-obsolete 'charset-list "use variable `charset-list'." "23.1")
478 (define-obsolete-function-alias 'char-valid-p 'characterp "23.1")
480 (defun generic-char-p (char)
481 "Always return nil. This is provided for backward compatibility."
483 (make-obsolete 'generic-char-p "generic characters no longer exist." "23.1")
485 (defun make-char-internal (charset-id &optional code1 code2)
486 (let ((charset (aref emacs-mule-charset-table charset-id)))
488 (error "Invalid Emacs-mule charset ID: %d" charset-id))
489 (make-char charset code1 code2)))
491 ;; Save the ASCII case table in case we need it later. Some locales
492 ;; (such as Turkish) modify the case behavior of ASCII characters,
493 ;; which can interfere with networking code that uses ASCII strings.
495 (defvar ascii-case-table
496 ;; Code copied from copy-case-table to avoid requiring case-table.el
497 (let ((tbl (copy-sequence (standard-case-table)))
498 (up (char-table-extra-slot (standard-case-table) 0)))
499 (if up (set-char-table-extra-slot tbl 0 (copy-sequence up)))
500 (set-char-table-extra-slot tbl 1 nil)
501 (set-char-table-extra-slot tbl 2 nil)
503 "Case table for the ASCII character set.")
505 ;; Coding system stuff
507 ;; Coding system is a symbol that has been defined by the function
508 ;; `define-coding-system'.
510 (defconst coding-system-iso-2022-flags
528 "List of symbols that control ISO-2022 encoder/decoder.
530 The value of the `:flags' attribute in the argument of the function
531 `define-coding-system' must be one of them.
533 If `long-form' is specified, use a long designation sequence on
534 encoding for the charsets `japanese-jisx0208-1978', `chinese-gb2312',
535 and `japanese-jisx0208'. The long designation sequence doesn't
536 conform to ISO 2022, but is used by such coding systems as
539 If `ascii-at-eol' is specified, designate ASCII to g0 at end of line
542 If `ascii-at-cntl' is specified, designate ASCII to g0 before control
543 codes and SPC on encoding.
545 If `7-bit' is specified, use 7-bit code only on encoding.
547 If `locking-shift' is specified, decode locking-shift code correctly
548 on decoding, and use locking-shift to invoke a graphic element on
551 If `single-shift' is specified, decode single-shift code correctly on
552 decoding, and use single-shift to invoke a graphic element on encoding.
554 If `designation' is specified, decode designation code correctly on
555 decoding, and use designation to designate a charset to a graphic
558 If `revision' is specified, produce an escape sequence to specify
559 revision number of a charset on encoding. Such an escape sequence is
560 always correctly decoded on decoding.
562 If `direction' is specified, decode ISO6429's code for specifying
563 direction correctly, and produce the code on encoding.
565 If `init-at-bol' is specified, on encoding, it is assumed that
566 invocation and designation statuses are reset at each beginning of
567 line even if `ascii-at-eol' is not specified; thus no codes for
568 resetting them are produced.
570 If `safe' is specified, on encoding, characters not supported by a
571 coding are replaced with `?'.
573 If `latin-extra' is specified, the code-detection routine assumes that a
574 code specified in `latin-extra-code-table' (which see) is valid.
576 If `composition' is specified, an escape sequence to specify
577 composition sequence is correctly decoded on decoding, and is produced
580 If `euc-tw-shift' is specified, the EUC-TW specific shifting code is
581 correctly decoded on decoding, and is produced on encoding.
583 If `use-roman' is specified, JIS0201-1976-Roman is designated instead
586 If `use-oldjis' is specified, JIS0208-1976 is designated instead of
589 (defun define-coding-system (name docstring &rest props)
590 "Define NAME (a symbol) as a coding system with DOCSTRING and attributes.
591 The remaining arguments must come in pairs ATTRIBUTE VALUE. ATTRIBUTE
594 The following attributes have special meanings. Those labeled as
595 \"(required)\" should not be omitted.
597 `:mnemonic' (required)
599 VALUE is a character to display on mode line for the coding system.
601 `:coding-type' (required)
603 VALUE must be one of `charset', `utf-8', `utf-16', `iso-2022',
604 `emacs-mule', `shift-jis', `ccl', `raw-text', `undecided'.
608 VALUE is the EOL (end-of-line) format of the coding system. It must be
609 one of `unix', `dos', `mac'. The symbol `unix' means Unix-like EOL
610 \(i.e. single LF), `dos' means DOS-like EOL \(i.e. sequence of CR LF),
611 and `mac' means Mac-like EOL \(i.e. single CR). If omitted, on
612 decoding by the coding system, Emacs automatically detects the EOL
613 format of the source text.
617 VALUE must be a list of charsets supported by the coding system. On
618 encoding by the coding system, if a character belongs to multiple
619 charsets in the list, a charset that comes earlier in the list is
620 selected. If `:coding-type' is `iso-2022', VALUE may be `iso-2022',
621 which indicates that the coding system supports all ISO-2022 based
622 charsets. If `:coding-type' is `emacs-mule', VALUE may be
623 `emacs-mule', which indicates that the coding system supports all
624 charsets that have the `:emacs-mule-id' property.
626 `:ascii-compatible-p'
628 If VALUE is non-nil, the coding system decodes all 7-bit bytes into
629 the corresponding ASCII characters, and encodes all ASCII characters
630 back to the corresponding 7-bit bytes. VALUE defaults to nil.
632 `:decode-translation-table'
634 VALUE must be a translation table to use on decoding.
636 `:encode-translation-table'
638 VALUE must be a translation table to use on encoding.
640 `:post-read-conversion'
642 VALUE must be a function to call after some text is inserted and
643 decoded by the coding system itself and before any functions in
644 `after-insert-functions' are called. The arguments to this function
645 are the same as those of a function in `after-insert-file-functions',
646 i.e. LENGTH of the text to be decoded with point at the head of it,
647 and the function should leave point unchanged.
649 `:pre-write-conversion'
651 VALUE must be a function to call after all functions in
652 `write-region-annotate-functions' and `buffer-file-format' are called,
653 and before the text is encoded by the coding system itself. The
654 arguments to this function are the same as those of a function in
655 `write-region-annotate-functions'.
659 VALUE must be a character. On encoding, a character not supported by
660 the coding system is replaced with VALUE.
664 VALUE non-nil means that visiting a file with the coding system
665 results in a unibyte buffer.
669 VALUE must be `unix', `dos', `mac'. The symbol `unix' means Unix-like
670 EOL (LF), `dos' means DOS-like EOL (CRLF), and `mac' means Mac-like
671 EOL (CR). If omitted, on decoding, the coding system detects EOL
672 format automatically, and on encoding, uses Unix-like EOL.
676 VALUE must be a symbol whose name is that of a MIME charset converted
679 `:mime-text-unsuitable'
681 VALUE non-nil means the `:mime-charset' property names a charset which
682 is unsuitable for the top-level media type \"text\".
686 VALUE must be a list of symbols that control the ISO-2022 converter.
687 Each must be a member of the list `coding-system-iso-2022-flags'
688 \(which see). This attribute has a meaning only when `:coding-type'
693 VALUE must be a vector [G0-USAGE G1-USAGE G2-USAGE G3-USAGE].
694 GN-USAGE specifies the usage of graphic register GN as follows.
696 If it is nil, no charset can be designated to GN.
698 If it is a charset, the charset is initially designated to GN, and
699 never used by the other charsets.
701 If it is a list, the elements must be charsets, nil, 94, or 96. GN
702 can be used by all the listed charsets. If the list contains 94, any
703 iso-2022 charset whose code-space ranges are 94 long can be designated
704 to GN. If the list contains 96, any charsets whose whose ranges are
705 96 long can be designated to GN. If the first element is a charset,
706 that charset is initially designated to GN.
708 This attribute has a meaning only when `:coding-type' is `iso-2022'.
712 This attributes specifies whether the coding system uses a `byte order
713 mark'. VALUE must be nil, t, or cons of coding systems whose
714 `:coding-type' is `utf-16' or `utf-8'.
716 If the value is nil, on decoding, don't treat the first two-byte as
717 BOM, and on encoding, don't produce BOM bytes.
719 If the value is t, on decoding, skip the first two-byte as BOM, and on
720 encoding, produce BOM bytes accoding to the value of `:endian'.
722 If the value is cons, on decoding, check the first two-byte. If they
723 are 0xFE 0xFF, use the car part coding system of the value. If they
724 are 0xFF 0xFE, use the cdr part coding system of the value.
725 Otherwise, treat them as bytes for a normal character. On encoding,
726 produce BOM bytes accoding to the value of `:endian'.
728 This attribute has a meaning only when `:coding-type' is `utf-16' or
733 VALUE must be `big' or `little' specifying big-endian and
734 little-endian respectively. The default value is `big'.
736 This attribute has a meaning only when `:coding-type' is `utf-16'.
740 VALUE is a symbol representing the registered CCL program used for
741 decoding. This attribute has a meaning only when `:coding-type' is
746 VALUE is a symbol representing the registered CCL program used for
747 encoding. This attribute has a meaning only when `:coding-type' is
749 (let* ((common-attrs (mapcar 'list
754 :decode-translation-table
755 :encode-translation-table
756 :post-read-conversion
757 :pre-write-conversion
762 (coding-type (plist-get props :coding-type))
763 (spec-attrs (mapcar 'list
764 (cond ((eq coding-type 'iso-2022)
769 ((eq coding-type 'utf-8)
771 ((eq coding-type 'utf-16)
774 ((eq coding-type 'ccl)
779 (dolist (slot common-attrs)
780 (setcdr slot (plist-get props (car slot))))
782 (dolist (slot spec-attrs)
783 (setcdr slot (plist-get props (car slot))))
785 (if (eq coding-type 'iso-2022)
786 (let ((designation (plist-get props :designation))
787 (flags (plist-get props :flags))
788 (initial (make-vector 4 nil))
789 (reg-usage (cons 4 4))
792 (setq elt (aref designation i))
793 (cond ((charsetp elt)
795 (setq request (cons (cons elt i) request)))
797 (aset initial i (car elt))
798 (if (charsetp (car elt))
799 (setq request (cons (cons (car elt) i) request)))
800 (dolist (e (cdr elt))
802 (setq request (cons (cons e i) request)))
804 (setcar reg-usage i))
806 (setcdr reg-usage i))
809 (setcdr reg-usage i)))))))
810 (setcdr (assq :initial spec-attrs) initial)
811 (setcdr (assq :reg-usage spec-attrs) reg-usage)
812 (setcdr (assq :request spec-attrs) request)
814 ;; Change :flags value from a list to a bit-mask.
817 (dolist (elt coding-system-iso-2022-flags)
819 (setq bits (logior bits (lsh 1 i))))
821 (setcdr (assq :flags spec-attrs) bits))))
823 ;; Add :name and :docstring properties to PROPS.
825 (cons :name (cons name (cons :docstring (cons (purecopy docstring)
827 (setcdr (assq :plist common-attrs) props)
828 (apply 'define-coding-system-internal
829 name (mapcar 'cdr (append common-attrs spec-attrs)))))
831 (defun coding-system-doc-string (coding-system)
832 "Return the documentation string for CODING-SYSTEM."
833 (plist-get (coding-system-plist coding-system) :docstring))
835 (defun coding-system-mnemonic (coding-system)
836 "Return the mnemonic character of CODING-SYSTEM.
837 The mnemonic character of a coding system is used in mode line to
838 indicate the coding system. If CODING-SYSTEM is nil, return ?=."
839 (plist-get (coding-system-plist coding-system) :mnemonic))
841 (defun coding-system-type (coding-system)
842 "Return the coding type of CODING-SYSTEM.
843 A coding type is a symbol indicating the encoding method of CODING-SYSTEM.
844 See the function `define-coding-system' for more detail."
845 (plist-get (coding-system-plist coding-system) :coding-type))
847 (defun coding-system-charset-list (coding-system)
848 "Return list of charsets supported by CODING-SYSTEM.
849 If CODING-SYSTEM supports all ISO-2022 charsets, return `iso-2022'.
850 If CODING-SYSTEM supports all emacs-mule charsets, return `emacs-mule'."
851 (plist-get (coding-system-plist coding-system) :charset-list))
853 (defun coding-system-category (coding-system)
854 "Return a category symbol of CODING-SYSTEM."
855 (plist-get (coding-system-plist coding-system) :category))
857 (defun coding-system-get (coding-system prop)
858 "Extract a value from CODING-SYSTEM's property list for property PROP.
859 For compatibility with Emacs 20/21, this accepts old-style symbols
860 like `mime-charset' as well as the current style like `:mime-charset'."
861 (or (plist-get (coding-system-plist coding-system) prop)
862 (if (not (keywordp prop))
863 ;; For backward compatibility.
864 (if (eq prop 'ascii-incompatible)
865 (not (plist-get (coding-system-plist coding-system)
866 :ascii-compatible-p))
867 (plist-get (coding-system-plist coding-system)
868 (intern (concat ":" (symbol-name prop))))))))
870 (defun coding-system-eol-type-mnemonic (coding-system)
871 "Return the string indicating end-of-line format of CODING-SYSTEM."
872 (let* ((eol-type (coding-system-eol-type coding-system))
873 (val (cond ((eq eol-type 0) eol-mnemonic-unix)
874 ((eq eol-type 1) eol-mnemonic-dos)
875 ((eq eol-type 2) eol-mnemonic-mac)
876 (t eol-mnemonic-undecided))))
879 (char-to-string val))))
881 (defun coding-system-lessp (x y)
882 (cond ((eq x 'no-conversion) t)
883 ((eq y 'no-conversion) nil)
884 ((eq x 'emacs-mule) t)
885 ((eq y 'emacs-mule) nil)
886 ((eq x 'undecided) t)
887 ((eq y 'undecided) nil)
888 (t (let ((c1 (coding-system-mnemonic x))
889 (c2 (coding-system-mnemonic y)))
890 (or (< (downcase c1) (downcase c2))
891 (and (not (> (downcase c1) (downcase c2)))
894 (defun coding-system-equal (coding-system-1 coding-system-2)
895 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
896 Two coding systems are identical if both symbols are equal
897 or one is an alias of the other."
898 (or (eq coding-system-1 coding-system-2)
899 (and (equal (coding-system-plist coding-system-1)
900 (coding-system-plist coding-system-2))
901 (let ((eol-type-1 (coding-system-eol-type coding-system-1))
902 (eol-type-2 (coding-system-eol-type coding-system-2)))
903 (or (eq eol-type-1 eol-type-2)
904 (and (vectorp eol-type-1) (vectorp eol-type-2)))))))
906 (defun add-to-coding-system-list (coding-system)
907 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
908 (if (or (null coding-system-list)
909 (coding-system-lessp coding-system (car coding-system-list)))
910 (setq coding-system-list (cons coding-system coding-system-list))
911 (let ((len (length coding-system-list))
912 mid (tem coding-system-list))
914 (setq mid (nthcdr (/ len 2) tem))
915 (if (coding-system-lessp (car mid) coding-system)
917 len (- len (/ len 2)))
918 (setq len (/ len 2))))
919 (setcdr tem (cons coding-system (cdr tem))))))
921 (defun coding-system-list (&optional base-only)
922 "Return a list of all existing non-subsidiary coding systems.
923 If optional arg BASE-ONLY is non-nil, only base coding systems are
924 listed. The value doesn't include subsidiary coding systems which are
925 made from bases and aliases automatically for various end-of-line
926 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
928 (dolist (coding coding-system-list)
929 (if (eq (coding-system-base coding) coding)
931 (setq codings (cons coding codings))
932 (dolist (alias (coding-system-aliases coding))
933 (setq codings (cons alias codings))))))
936 (defconst char-coding-system-table nil
937 "It exists just for backward compatibility, and the value is always nil.")
938 (make-obsolete-variable 'char-coding-system-table nil "23.1")
940 (defun transform-make-coding-system-args (name type &optional doc-string props)
941 "For internal use only.
942 Transform XEmacs style args for `make-coding-system' to Emacs style.
943 Value is a list of transformed arguments."
944 (let ((mnemonic (string-to-char (or (plist-get props 'mnemonic) "?")))
945 (eol-type (plist-get props 'eol-type))
948 ((eq eol-type 'lf) (setq eol-type 'unix))
949 ((eq eol-type 'crlf) (setq eol-type 'dos))
950 ((eq eol-type 'cr) (setq eol-type 'mac)))
951 (if (setq tmp (plist-get props 'post-read-conversion))
952 (setq properties (plist-put properties 'post-read-conversion tmp)))
953 (if (setq tmp (plist-get props 'pre-write-conversion))
954 (setq properties (plist-put properties 'pre-write-conversion tmp)))
956 ((eq type 'shift-jis)
957 `(,name 1 ,mnemonic ,doc-string () ,properties ,eol-type))
958 ((eq type 'iso2022) ; This is not perfect.
959 (if (plist-get props 'escape-quoted)
960 (error "escape-quoted is not supported: %S"
961 `(,name ,type ,doc-string ,props)))
962 (let ((g0 (plist-get props 'charset-g0))
963 (g1 (plist-get props 'charset-g1))
964 (g2 (plist-get props 'charset-g2))
965 (g3 (plist-get props 'charset-g3))
968 (eq (cadr (assoc 'latin-jisx0201
969 (plist-get props 'input-charset-conversion)))
971 (eq (cadr (assoc 'ascii
972 (plist-get props 'output-charset-conversion)))
976 (eq (cadr (assoc 'japanese-jisx0208-1978
977 (plist-get props 'input-charset-conversion)))
979 (eq (cadr (assoc 'japanese-jisx0208
980 (plist-get props 'output-charset-conversion)))
981 'japanese-jisx0208-1978))))
983 (if (plist-get props 'force-g0-on-output)
987 (if (plist-get props 'force-g1-on-output)
991 (if (plist-get props 'force-g2-on-output)
995 (if (plist-get props 'force-g3-on-output)
998 `(,name 2 ,mnemonic ,doc-string
1000 ,(plist-get props 'short)
1001 ,(not (plist-get props 'no-ascii-eol))
1002 ,(not (plist-get props 'no-ascii-cntl))
1003 ,(plist-get props 'seven)
1005 ,(not (plist-get props 'lock-shift))
1008 ,(plist-get props 'no-iso6429)
1010 ,properties ,eol-type)))
1012 `(,name 3 ,mnemonic ,doc-string () ,properties ,eol-type))
1014 `(,name 4 ,mnemonic ,doc-string
1015 (,(plist-get props 'decode) . ,(plist-get props 'encode))
1016 ,properties ,eol-type))
1018 (error "unsupported XEmacs style make-coding-style arguments: %S"
1019 `(,name ,type ,doc-string ,props))))))
1021 (defun make-coding-system (coding-system type mnemonic doc-string
1026 "Define a new coding system CODING-SYSTEM (symbol).
1027 This function is provided for backward compatibility."
1028 ;; For compatibility with XEmacs, we check the type of TYPE. If it
1029 ;; is a symbol, perhaps, this function is called with XEmacs-style
1030 ;; arguments. Here, try to transform that kind of arguments to
1033 (let ((args (transform-make-coding-system-args coding-system type
1034 mnemonic doc-string)))
1035 (setq coding-system (car args)
1037 mnemonic (nth 2 args)
1038 doc-string (nth 3 args)
1040 properties (nth 5 args)
1041 eol-type (nth 6 args))))
1044 (cond ((eq type 0) 'emacs-mule)
1045 ((eq type 1) 'shift-jis)
1046 ((eq type 2) 'iso2022)
1049 ((eq type 5) 'raw-text)
1051 (error "Invalid coding system type: %s" type))))
1054 (let ((plist nil) key)
1055 (dolist (elt properties)
1056 (setq key (car elt))
1057 (cond ((eq key 'post-read-conversion)
1058 (setq key :post-read-conversion))
1059 ((eq key 'pre-write-conversion)
1060 (setq key :pre-write-conversion))
1061 ((eq key 'translation-table-for-decode)
1062 (setq key :decode-translation-table))
1063 ((eq key 'translation-table-for-encode)
1064 (setq key :encode-translation-table))
1065 ((eq key 'safe-charsets)
1066 (setq key :charset-list))
1067 ((eq key 'mime-charset)
1068 (setq key :mime-charset))
1069 ((eq key 'valid-codes)
1070 (setq key :valids)))
1071 (setq plist (plist-put plist key (cdr elt))))
1073 (setq properties (plist-put properties :mnemonic mnemonic))
1074 (plist-put properties :coding-type type)
1075 (cond ((eq eol-type 0) (setq eol-type 'unix))
1076 ((eq eol-type 1) (setq eol-type 'dos))
1077 ((eq eol-type 2) (setq eol-type 'mac))
1078 ((vectorp eol-type) (setq eol-type nil)))
1079 (plist-put properties :eol-type eol-type)
1083 (plist-put properties :flags
1084 (list (and (or (consp (nth 0 flags))
1085 (consp (nth 1 flags))
1086 (consp (nth 2 flags))
1087 (consp (nth 3 flags))) 'designation)
1088 (or (nth 4 flags) 'long-form)
1089 (and (nth 5 flags) 'ascii-at-eol)
1090 (and (nth 6 flags) 'ascii-at-cntl)
1091 (and (nth 7 flags) '7-bit)
1092 (and (nth 8 flags) 'locking-shift)
1093 (and (nth 9 flags) 'single-shift)
1094 (and (nth 10 flags) 'use-roman)
1095 (and (nth 11 flags) 'use-oldjis)
1096 (or (nth 12 flags) 'direction)
1097 (and (nth 13 flags) 'init-at-bol)
1098 (and (nth 14 flags) 'designate-at-bol)
1099 (and (nth 15 flags) 'safe)
1100 (and (nth 16 flags) 'latin-extra)))
1101 (plist-put properties :designation
1102 (let ((vec (make-vector 4 nil)))
1104 (let ((spec (nth i flags)))
1106 (aset vec i '(94 96))
1110 (setq spec (append (delq t spec) '(94 96))))
1111 (aset vec i spec))))))
1115 (plist-put properties :ccl-decoder (car flags))
1116 (plist-put properties :ccl-encoder (cdr flags))))
1118 (apply 'define-coding-system coding-system doc-string properties))
1120 (make-obsolete 'make-coding-system 'define-coding-system "23.1")
1122 (defun merge-coding-systems (first second)
1123 "Fill in any unspecified aspects of coding system FIRST from SECOND.
1124 Return the resulting coding system."
1125 (let ((base (coding-system-base second))
1126 (eol (coding-system-eol-type second)))
1127 ;; If FIRST doesn't specify text conversion, merge with that of SECOND.
1128 (if (eq (coding-system-base first) 'undecided)
1129 (setq first (coding-system-change-text-conversion first base)))
1130 ;; If FIRST doesn't specify eol conversion, merge with that of SECOND.
1131 (if (and (vectorp (coding-system-eol-type first))
1132 (numberp eol) (>= eol 0) (<= eol 2))
1133 (setq first (coding-system-change-eol-conversion
1137 (defun autoload-coding-system (symbol form)
1138 "Define SYMBOL as a coding-system that is defined on demand.
1140 FORM is a form to evaluate to define the coding-system."
1141 (put symbol 'coding-system-define-form form)
1142 (setq coding-system-alist (cons (list (symbol-name symbol))
1143 coding-system-alist))
1144 (dolist (elt '("-unix" "-dos" "-mac"))
1145 (let ((name (concat (symbol-name symbol) elt)))
1146 (put (intern name) 'coding-system-define-form form)
1147 (setq coding-system-alist (cons (list name) coding-system-alist)))))
1149 ;; This variable is set in these three cases:
1150 ;; (1) A file is read by a coding system specified explicitly.
1151 ;; after-insert-file-set-coding sets the car of this value to
1152 ;; coding-system-for-read, and sets the cdr to nil.
1153 ;; (2) A buffer is saved.
1154 ;; After writing, basic-save-buffer-1 sets the car of this value
1155 ;; to last-coding-system-used.
1156 ;; (3) set-buffer-file-coding-system is called.
1157 ;; The cdr of this value is set to the specified coding system.
1158 ;; This variable is used for decoding in revert-buffer and encoding in
1159 ;; select-safe-coding-system.
1160 (defvar buffer-file-coding-system-explicit nil
1161 "The file coding system explicitly specified for the current buffer.
1162 The value is a cons of coding systems for reading (decoding) and
1164 Internal use only.")
1165 (make-variable-buffer-local 'buffer-file-coding-system-explicit)
1166 (put 'buffer-file-coding-system-explicit 'permanent-local t)
1168 (defun set-buffer-file-coding-system (coding-system &optional force nomodify)
1169 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1170 This means that when you save the buffer, it will be converted
1171 according to CODING-SYSTEM. For a list of possible values of
1172 CODING-SYSTEM, use \\[list-coding-systems].
1174 If CODING-SYSTEM leaves the text conversion unspecified, or if it leaves
1175 the end-of-line conversion unspecified, FORCE controls what to do.
1176 If FORCE is nil, get the unspecified aspect (or aspects) from the buffer's
1177 previous `buffer-file-coding-system' value (if it is specified there).
1178 Otherwise, leave it unspecified.
1180 This marks the buffer modified so that the succeeding \\[save-buffer]
1181 surely saves the buffer with CODING-SYSTEM. From a program, if you
1182 don't want to mark the buffer modified, specify t for NOMODIFY.
1183 If you know exactly what coding system you want to use,
1184 just set the variable `buffer-file-coding-system' directly."
1185 (interactive "zCoding system for saving file (default nil): \nP")
1186 (check-coding-system coding-system)
1187 (if (and coding-system buffer-file-coding-system (null force))
1189 (merge-coding-systems coding-system buffer-file-coding-system)))
1190 (setq buffer-file-coding-system coding-system)
1191 (if buffer-file-coding-system-explicit
1192 (setcdr buffer-file-coding-system-explicit coding-system)
1193 (setq buffer-file-coding-system-explicit (cons nil coding-system)))
1194 ;; This is in case of an explicit call. Normally, `normal-mode' and
1195 ;; `set-buffer-major-mode-hook' take care of setting the table.
1196 (if (fboundp 'ucs-set-table-for-input) ; don't lose when building
1197 (ucs-set-table-for-input))
1199 (set-buffer-modified-p t))
1200 (force-mode-line-update))
1202 (defun revert-buffer-with-coding-system (coding-system &optional force)
1203 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1204 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1206 If CODING-SYSTEM leaves the text conversion unspecified, or if it leaves
1207 the end-of-line conversion unspecified, FORCE controls what to do.
1208 If FORCE is nil, get the unspecified aspect (or aspects) from the buffer's
1209 previous `buffer-file-coding-system' value (if it is specified there).
1210 Otherwise, determine it from the file contents as usual for visiting a file."
1211 (interactive "zCoding system for visited file (default nil): \nP")
1212 (check-coding-system coding-system)
1213 (if (and coding-system buffer-file-coding-system (null force))
1215 (merge-coding-systems coding-system buffer-file-coding-system)))
1216 (let ((coding-system-for-read coding-system))
1219 (defun set-file-name-coding-system (coding-system)
1220 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1221 It actually just set the variable `file-name-coding-system' (which see)
1223 (interactive "zCoding system for file names (default nil): ")
1224 (check-coding-system coding-system)
1225 (if (and coding-system
1226 (not (coding-system-get coding-system :ascii-compatible-p))
1227 (not (coding-system-get coding-system :suitable-for-file-name)))
1228 (error "%s is not suitable for file names" coding-system))
1229 (setq file-name-coding-system coding-system))
1231 (defvar default-terminal-coding-system nil
1232 "Default value for the terminal coding system.
1233 This is normally set according to the selected language environment.
1234 See also the command `set-terminal-coding-system'.")
1236 (defun set-terminal-coding-system (coding-system &optional terminal)
1237 "Set coding system of terminal output to CODING-SYSTEM.
1238 All text output to TERMINAL will be encoded
1239 with the specified coding system.
1241 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1242 The default is determined by the selected language environment
1243 or by the previous use of this command.
1245 TERMINAL may be a terminal object, a frame, or nil for the
1246 selected frame's terminal. The setting has no effect on
1247 graphical terminals."
1249 (list (let ((default (if (and (not (terminal-coding-system))
1250 default-terminal-coding-system)
1251 default-terminal-coding-system)))
1253 (format "Coding system for terminal display (default %s): "
1256 (if (and (not coding-system)
1257 (not (terminal-coding-system)))
1258 (setq coding-system default-terminal-coding-system))
1260 (setq default-terminal-coding-system coding-system))
1261 (set-terminal-coding-system-internal coding-system terminal)
1262 (redraw-frame (selected-frame)))
1264 (defvar default-keyboard-coding-system nil
1265 "Default value of the keyboard coding system.
1266 This is normally set according to the selected language environment.
1267 See also the command `set-keyboard-coding-system'.")
1269 (defun set-keyboard-coding-system (coding-system &optional terminal)
1270 "Set coding system for keyboard input on TERMINAL to CODING-SYSTEM.
1272 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1273 The default is determined by the selected language environment
1274 or by the previous use of this command.
1276 If CODING-SYSTEM is nil or the coding-type of CODING-SYSTEM is
1277 `raw-text', the decoding of keyboard input is disabled.
1279 TERMINAL may be a terminal object, a frame, or nil for the
1280 selected frame's terminal. The setting has no effect on
1281 graphical terminals."
1283 (list (let* ((coding (keyboard-coding-system nil))
1284 (default (if (eq (coding-system-type coding) 'raw-text)
1285 default-keyboard-coding-system)))
1287 (format "Coding system for keyboard input (default %s): "
1290 (let ((coding-type (coding-system-type coding-system))
1292 (terminal-parameter terminal 'keyboard-coding-saved-meta-mode)))
1293 (if (not (eq coding-type 'raw-text))
1295 (if (not (or (coding-system-get coding-system :suitable-for-keyboard)
1296 (coding-system-get coding-system :ascii-compatible-p)))
1297 (error "Unsuitable coding system for keyboard: %s" coding-system))
1298 (cond ((memq coding-type '(charset utf-8 shift-jis big5 ccl))
1299 (setq accept-8-bit t))
1300 ((eq coding-type 'iso-2022)
1301 (let ((flags (coding-system-get coding-system :flags)))
1302 (or (memq '7-bit flags)
1303 (setq accept-8-bit t))))
1305 (error "Unsupported coding system for keyboard: %s"
1309 (set-terminal-parameter terminal
1310 'keyboard-coding-saved-meta-mode
1311 (cons (nth 2 (current-input-mode))
1313 (set-input-meta-mode 8))
1314 ;; Avoid end-of-line conversion.
1316 (coding-system-change-eol-conversion coding-system 'unix)))
1318 (when saved-meta-mode
1319 (set-input-meta-mode (car saved-meta-mode))
1320 (set-terminal-parameter terminal
1321 'keyboard-coding-saved-meta-mode
1323 (set-keyboard-coding-system-internal coding-system terminal)
1324 (setq keyboard-coding-system coding-system))
1326 (defcustom keyboard-coding-system nil
1327 "Specify coding system for keyboard input.
1328 If you set this on a terminal which can't distinguish Meta keys from
1329 8-bit characters, you will have to use ESC to type Meta characters.
1330 See Info node `Terminal Coding' and Info node `Unibyte Mode'.
1332 On non-windowing terminals, this is set from the locale by default.
1334 Setting this variable directly does not take effect;
1335 use either \\[customize] or \\[set-keyboard-coding-system]."
1336 :type '(coding-system :tag "Coding system")
1337 :link '(info-link "(emacs)Terminal Coding")
1338 :link '(info-link "(emacs)Unibyte Mode")
1339 :set (lambda (symbol value)
1340 ;; Don't load encoded-kb unnecessarily.
1341 (if (or value (boundp 'encoded-kbd-setup-display))
1342 (set-keyboard-coding-system value)
1343 (set-default 'keyboard-coding-system nil))) ; must initialize
1348 (defun set-buffer-process-coding-system (decoding encoding)
1349 "Set coding systems for the process associated with the current buffer.
1350 DECODING is the coding system to be used to decode input from the process,
1351 ENCODING is the coding system to be used to encode output to the process.
1353 For a list of possible coding systems, use \\[list-coding-systems]."
1355 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1356 (let ((proc (get-buffer-process (current-buffer))))
1358 (error "No process")
1359 (check-coding-system decoding)
1360 (check-coding-system encoding)
1361 (set-process-coding-system proc decoding encoding)))
1362 (force-mode-line-update))
1364 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
1366 (defun set-selection-coding-system (coding-system)
1367 "Make CODING-SYSTEM used for communicating with other X clients.
1368 When sending or receiving text via cut_buffer, selection, and clipboard,
1369 the text is encoded or decoded by CODING-SYSTEM."
1370 (interactive "zCoding system for X selection: ")
1371 (check-coding-system coding-system)
1372 (setq selection-coding-system coding-system))
1374 ;; Coding system lastly specified by the command
1375 ;; set-next-selection-coding-system.
1376 (defvar last-next-selection-coding-system nil)
1378 (defun set-next-selection-coding-system (coding-system)
1379 "Use CODING-SYSTEM for next communication with other window system clients.
1380 This setting is effective for the next communication only."
1382 (list (read-coding-system
1383 (if last-next-selection-coding-system
1384 (format "Coding system for the next selection (default %S): "
1385 last-next-selection-coding-system)
1386 "Coding system for the next selection: ")
1387 last-next-selection-coding-system)))
1389 (setq last-next-selection-coding-system coding-system)
1390 (setq coding-system last-next-selection-coding-system))
1391 (check-coding-system coding-system)
1393 (setq next-selection-coding-system coding-system))
1395 (defun set-coding-priority (arg)
1396 "Set priority of coding categories according to ARG.
1397 ARG is a list of coding categories ordered by priority.
1399 This function is provided for backward compatibility."
1400 (apply 'set-coding-system-priority
1401 (mapcar #'(lambda (x) (symbol-value x)) arg)))
1402 (make-obsolete 'set-coding-priority 'set-coding-system-priority "23.1")
1406 (defvar ctext-non-standard-encodings-alist
1408 '(("big5-0" big5 2 big5)
1409 ("ISO8859-14" iso-8859-14 1 latin-iso8859-14)
1410 ("ISO8859-15" iso-8859-15 1 latin-iso8859-15)
1411 ("gbk-0" gbk 2 chinese-gbk)
1412 ("koi8-r" koi8-r 1 koi8-r)
1413 ("microsoft-cp1251" windows-1251 1 windows-1251)))
1414 "Alist of non-standard encoding names vs the corresponding usages in CTEXT.
1416 It controls how extended segments of a compound text are handled
1417 by the coding system `compound-text-with-extensions'.
1419 Each element has the form (ENCODING-NAME CODING-SYSTEM N-OCTET CHARSET).
1421 ENCODING-NAME is an encoding name of an \"extended segment\".
1423 CODING-SYSTEM is the coding-system to encode (or decode) the
1424 characters into (or from) the extended segment.
1426 N-OCTET is the number of octets (bytes) that encodes a character
1427 in the segment. It can be 0 (meaning the number of octets per
1428 character is variable), 1, 2, 3, or 4.
1430 CHARSET is a character set containing characters that are encoded
1431 in the segment. It can be a list of character sets.
1433 On decoding CTEXT, all encoding names listed here are recognized.
1435 On encoding CTEXT, encoding names in the variable
1436 `ctext-non-standard-encodings' (which see) and in the information
1437 listed for the current language environment under the key
1438 `ctext-non-standard-encodings' are used.")
1440 (defvar ctext-non-standard-encodings nil
1441 "List of non-standard encoding names used in extended segments of CTEXT.
1442 Each element must be one of the names listed in the variable
1443 `ctext-non-standard-encodings-alist' (which see).")
1445 (defvar ctext-non-standard-encodings-regexp
1447 (string-to-multibyte
1449 ;; For non-standard encodings.
1450 "\\(\e%/[0-4][\200-\377][\200-\377]\\([^\002]+\\)\002\\)"
1452 ;; For UTF-8 encoding.
1453 "\\(\e%G[^\e]*\e%@\\)"))))
1455 ;; Functions to support "Non-Standard Character Set Encodings" defined
1456 ;; by the COMPOUND-TEXT spec. They also support "The UTF-8 encoding"
1457 ;; described in the section 7 of the documentation of COMPOUND-TEXT
1458 ;; distributed with XFree86.
1460 (defun ctext-post-read-conversion (len)
1461 "Decode LEN characters encoded as Compound Text with Extended Segments."
1462 ;; We don't need the following because it is expected that this
1463 ;; function is mainly used for decoding X selection which is not
1465 ;;(buffer-disable-undo) ; minimize consing due to insertions and deletions
1468 (narrow-to-region (point) (+ (point) len))
1469 (let ((case-fold-search nil)
1470 last-coding-system-used
1472 (decode-coding-region (point-min) (point-max) 'ctext)
1473 (while (re-search-forward ctext-non-standard-encodings-regexp
1475 (setq pos (match-beginning 0))
1476 (if (match-beginning 1)
1477 ;; ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
1478 (let* ((M (multibyte-char-to-unibyte (char-after (+ pos 4))))
1479 (L (multibyte-char-to-unibyte (char-after (+ pos 5))))
1480 (encoding (match-string 2))
1481 (encoding-info (assoc-string
1483 ctext-non-standard-encodings-alist t))
1484 (coding (if encoding-info
1485 (nth 1 encoding-info)
1486 (setq encoding (intern (downcase encoding)))
1487 (and (coding-system-p encoding)
1489 (setq bytes (- (+ (* (- M 128) 128) (- L 128))
1490 (- (point) (+ pos 6))))
1492 (delete-region pos (point))
1493 (forward-char bytes)
1494 (decode-coding-region (- (point) bytes) (point) coding)))
1495 ;; ESC % G --UTF-8-BYTES-- ESC % @
1497 (delete-region pos (+ pos 3))
1498 (decode-coding-region pos (point) 'utf-8))))
1499 (goto-char (point-min))
1500 (- (point-max) (point)))))
1502 (defvar ctext-standard-encodings
1503 '(ascii latin-jisx0201 katakana-jisx0201
1504 latin-iso8859-1 latin-iso8859-2 latin-iso8859-3 latin-iso8859-4
1505 greek-iso8859-7 arabic-iso8859-6 hebrew-iso8859-8 cyrillic-iso8859-5
1507 chinese-gb2312 japanese-jisx0208 korean-ksc5601)
1508 "List of approved standard encodings (i.e. charsets) of X's Compound Text.
1509 Coding-system `compound-text-with-extensions' encodes a character
1510 belonging to any of those charsets using the normal ISO2022
1511 designation sequence unless the current language environment or
1512 the variable `ctext-non-standard-encodings' decide to use an extended
1513 segment of CTEXT for that character. See also the documentation
1514 of `ctext-non-standard-encodings-alist'.")
1516 ;; Return an alist of CHARSET vs CTEXT-USAGE-INFO generated from
1517 ;; `ctext-non-standard-encodings' and a list specified by the key
1518 ;; `ctext-non-standard-encodings' for the currrent language
1519 ;; environment. CTEXT-USAGE-INFO is one of the element of
1520 ;; `ctext-non-standard-encodings-alist' or nil. In the former case, a
1521 ;; character in CHARSET is encoded using extended segment. In the
1522 ;; latter case, a character in CHARSET is encoded using normal ISO2022
1523 ;; designation sequence. If a character is not in any of CHARSETs, it
1524 ;; is encoded using UTF-8 encoding extention.
1526 (defun ctext-non-standard-encodings-table ()
1527 (let* ((table (append ctext-non-standard-encodings
1529 (get-language-info current-language-environment
1530 'ctext-non-standard-encodings))))
1534 (setq elt (car tail))
1535 (let* ((slot (assoc elt ctext-non-standard-encodings-alist))
1536 (charset (nth 3 slot)))
1537 (if (charsetp charset)
1538 (setcar tail (cons charset slot))
1539 (setcar tail (cons (car charset) slot))
1540 (dolist (cs (cdr charset))
1542 (cons (cons (car cs) slot) (cdr tail)))
1543 (setq tail (cdr tail))))
1544 (setq tail (cdr tail))))
1547 (defun ctext-pre-write-conversion (from to)
1548 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1550 If FROM is a string, generate a new temp buffer, insert the text,
1551 and convert it in the temporary buffer. Otherwise, convert
1554 ;; Setup a working buffer if necessary.
1555 (when (stringp from)
1556 (set-buffer (generate-new-buffer " *temp"))
1557 (set-buffer-multibyte (multibyte-string-p from))
1559 (setq from 1 to (point-max)))
1561 (narrow-to-region from to)
1562 (let ((encoding-table (ctext-non-standard-encodings-table))
1563 (charset-list ctext-standard-encodings)
1564 last-coding-system-used
1565 last-pos last-encoding-info
1566 encoding-info end-pos ch charset)
1567 (dolist (elt encoding-table)
1568 (push (car elt) charset-list))
1569 (goto-char (setq last-pos from))
1570 (setq end-pos (point-marker))
1571 (while (re-search-forward "[^\000-\177]+" nil t)
1572 ;; Found a sequence of non-ASCII characters.
1573 (setq last-pos (match-beginning 0)
1574 ch (char-after last-pos)
1575 charset (char-charset ch charset-list)
1578 (or (cdr (assq charset encoding-table))
1581 (set-marker end-pos (match-end 0))
1582 (goto-char (1+ last-pos))
1583 (while (marker-position end-pos)
1584 (if (< (point) end-pos)
1586 (setq charset (char-charset (following-char) charset-list)
1589 (or (cdr (assq charset encoding-table))
1593 (setq encoding-info nil)
1594 (set-marker end-pos nil))
1595 (unless (eq last-encoding-info encoding-info)
1596 (cond ((consp last-encoding-info)
1597 ;; Encode the previous range using an extended
1599 (let ((encoding-name (car last-encoding-info))
1600 (coding-system (nth 1 last-encoding-info))
1601 (noctets (nth 2 last-encoding-info))
1603 (encode-coding-region last-pos (point) coding-system)
1604 (setq len (+ (length encoding-name) 1
1605 (- (point) last-pos)))
1606 ;; According to the spec of CTEXT, it is not
1607 ;; necessary to produce this extra designation
1608 ;; sequence, but some buggy application
1609 ;; (e.g. crxvt-gb) requires it.
1612 (goto-char last-pos)
1613 (insert (format "\e%%/%d" noctets))
1614 (insert-byte (+ (/ len 128) 128) 1)
1615 (insert-byte (+ (% len 128) 128) 1)
1616 (insert encoding-name)
1618 ((eq last-encoding-info 'utf-8)
1619 ;; Encode the previous range using UTF-8 encoding
1621 (encode-coding-region last-pos (point) 'mule-utf-8)
1623 (goto-char last-pos)
1627 (put-text-property last-pos (point) 'charset charset)))
1628 (setq last-pos (point)
1629 last-encoding-info encoding-info))))
1630 (goto-char (point-min)))))
1631 ;; Must return nil, as build_annotations_2 expects that.
1636 (defcustom auto-coding-alist
1637 ;; .exe and .EXE are added to support archive-mode looking at DOS
1638 ;; self-extracting exe archives.
1639 (mapcar (lambda (arg) (cons (purecopy (car arg)) (cdr arg)))
1641 arc\\|zip\\|lzh\\|lha\\|zoo\\|[jew]ar\\|xpi\\|rar\\|\
1642 ARC\\|ZIP\\|LZH\\|LHA\\|ZOO\\|[JEW]AR\\|XPI\\|RAR\\)\\'"
1643 . no-conversion-multibyte)
1644 ("\\.\\(exe\\|EXE\\)\\'" . no-conversion)
1645 ("\\.\\(sx[dmicw]\\|odt\\|tar\\|tgz\\)\\'" . no-conversion)
1646 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|gpg\\)\\'" . no-conversion)
1647 ("\\.\\(jpe?g\\|png\\|gif\\|tiff?\\|p[bpgn]m\\)\\'" . no-conversion)
1648 ("\\.pdf\\'" . no-conversion)
1649 ("/#[^/]+#\\'" . emacs-mule)))
1650 "Alist of filename patterns vs corresponding coding systems.
1651 Each element looks like (REGEXP . CODING-SYSTEM).
1652 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1654 The settings in this alist take priority over `coding:' tags
1655 in the file (see the function `set-auto-coding')
1656 and the contents of `file-coding-system-alist'."
1659 :type '(repeat (cons (regexp :tag "File name regexp")
1660 (symbol :tag "Coding system"))))
1662 (defcustom auto-coding-regexp-alist
1663 (mapcar (lambda (arg) (cons (purecopy (car arg)) (cdr arg)))
1664 '(("\\`BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion)
1665 ("\\`\xFE\xFF" . utf-16be-with-signature)
1666 ("\\`\xFF\xFE" . utf-16le-with-signature)
1667 ("\\`\xEF\xBB\xBF" . utf-8-with-signature)
1668 ("\\`;ELC\024\0\0\0" . emacs-mule))) ; Emacs 20-compiled
1669 "Alist of patterns vs corresponding coding systems.
1670 Each element looks like (REGEXP . CODING-SYSTEM).
1671 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1673 The settings in this alist take priority over `coding:' tags
1674 in the file (see the function `set-auto-coding')
1675 and the contents of `file-coding-system-alist'."
1678 :type '(repeat (cons (regexp :tag "Regexp")
1679 (symbol :tag "Coding system"))))
1681 (defun auto-coding-regexp-alist-lookup (from to)
1682 "Lookup `auto-coding-regexp-alist' for the contents of the current buffer.
1683 The value is a coding system is specified for the region FROM and TO,
1687 (let ((alist auto-coding-regexp-alist)
1689 (while (and alist (not coding-system))
1690 (let ((regexp (car (car alist))))
1691 (if enable-multibyte-characters
1692 (setq regexp (string-to-multibyte regexp)))
1693 (if (re-search-forward regexp to t)
1694 (setq coding-system (cdr (car alist)))
1695 (setq alist (cdr alist)))))
1698 ;; See the bottom of this file for built-in auto coding functions.
1699 (defcustom auto-coding-functions '(sgml-xml-auto-coding-function
1700 sgml-html-meta-auto-coding-function)
1701 "A list of functions which attempt to determine a coding system.
1703 Each function in this list should be written to operate on the
1704 current buffer, but should not modify it in any way. The buffer
1705 will contain undecoded text of parts of the file. Each function
1706 should take one argument, SIZE, which says how many characters
1707 \(starting from point) it should look at.
1709 If one of these functions succeeds in determining a coding
1710 system, it should return that coding system. Otherwise, it
1713 If a file has a `coding:' tag, that takes precedence over these
1714 functions, so they won't be called at all."
1717 :type '(repeat function))
1719 (defvar set-auto-coding-for-load nil
1720 "Non-nil means look for `load-coding' property instead of `coding'.
1721 This is used for loading and byte-compiling Emacs Lisp files.")
1723 (defun auto-coding-alist-lookup (filename)
1724 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1725 (let ((alist auto-coding-alist)
1726 (case-fold-search (memq system-type '(windows-nt ms-dos cygwin)))
1728 (while (and alist (not coding-system))
1729 (if (string-match (car (car alist)) filename)
1730 (setq coding-system (cdr (car alist)))
1731 (setq alist (cdr alist))))
1734 (put 'enable-character-translation 'permanent-local t)
1735 (put 'enable-character-translation 'safe-local-variable 'booleanp)
1737 (defun find-auto-coding (filename size)
1738 "Find a coding system for a file FILENAME of which SIZE bytes follow point.
1739 These bytes should include at least the first 1k of the file
1740 and the last 3k of the file, but the middle may be omitted.
1742 The function checks FILENAME against the variable `auto-coding-alist'.
1743 If FILENAME doesn't match any entries in the variable, it checks the
1744 contents of the current buffer following point against
1745 `auto-coding-regexp-alist'. If no match is found, it checks for a
1746 `coding:' tag in the first one or two lines following point. If no
1747 `coding:' tag is found, it checks any local variables list in the last
1748 3K bytes out of the SIZE bytes. Finally, if none of these methods
1749 succeed, it checks to see if any function in `auto-coding-functions'
1752 If a coding system is specifed, the return value is a cons
1753 \(CODING . SOURCE), where CODING is the specified coding system and
1754 SOURCE is a symbol `auto-coding-alist', `auto-coding-regexp-alist',
1755 `:coding', or `auto-coding-functions' indicating by what CODING is
1756 specified. Note that the validity of CODING is not checked;
1757 it's the caller's responsibility to check it.
1759 If nothing is specified, the return value is nil."
1760 (or (let ((coding-system (auto-coding-alist-lookup filename)))
1762 (cons coding-system 'auto-coding-alist)))
1763 ;; Try using `auto-coding-regexp-alist'.
1764 (let ((coding-system (auto-coding-regexp-alist-lookup (point)
1767 (cons coding-system 'auto-coding-regexp-alist)))
1768 (let* ((case-fold-search t)
1769 (head-start (point))
1770 (head-end (+ head-start (min size 1024)))
1771 (tail-start (+ head-start (max (- size 3072) 0)))
1772 (tail-end (+ head-start size))
1773 coding-system head-found tail-found pos char-trans)
1774 ;; Try a short cut by searching for the string "coding:"
1775 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1776 (setq head-found (or (search-forward "coding:" head-end t)
1777 (search-forward "unibyte:" head-end t)
1778 (search-forward "enable-character-translation:"
1780 (if (and head-found (> head-found tail-start))
1781 ;; Head and tail are overlapped.
1782 (setq tail-found head-found)
1783 (goto-char tail-start)
1784 (setq tail-found (or (search-forward "coding:" tail-end t)
1785 (search-forward "unibyte:" tail-end t)
1786 (search-forward "enable-character-translation:"
1789 ;; At first check the head.
1791 (goto-char head-start)
1792 (setq head-end (set-auto-mode-1))
1793 (setq head-start (point))
1794 (when (and head-end (< head-found head-end))
1795 (goto-char head-start)
1796 (when (and set-auto-coding-for-load
1798 "\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1800 (setq coding-system 'raw-text))
1801 (when (and (not coding-system)
1803 "\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1805 (setq coding-system (intern (match-string 2))))
1806 (when (re-search-forward
1807 "\\(.*;\\)?[ \t]*enable-character-translation:[ \t]*\\([^ ;]+\\)"
1809 (setq char-trans (match-string 2)))))
1811 ;; If no coding: tag in the head, check the tail.
1812 ;; Here we must pay attention to the case that the end-of-line
1813 ;; is just "\r" and we can't use "^" nor "$" in regexp.
1814 (when (and tail-found (or (not coding-system) (not char-trans)))
1815 (goto-char tail-start)
1816 (re-search-forward "[\r\n]\^L" tail-end t)
1817 (if (re-search-forward
1818 "[\r\n]\\([^[\r\n]*\\)[ \t]*Local Variables:[ \t]*\\([^\r\n]*\\)[\r\n]"
1820 ;; The prefix is what comes before "local variables:" in its
1821 ;; line. The suffix is what comes after "local variables:"
1823 (let* ((prefix (regexp-quote (match-string 1)))
1824 (suffix (regexp-quote (match-string 2)))
1828 ;; N.B. without the \n below, the regexp can
1830 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1835 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1840 "[ \t]*enable-character-translation[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1843 (concat "[\r\n]" prefix "[ \t]*End *:[ \t]*" suffix
1846 (forward-char -1) ; skip back \r or \n.
1847 (re-search-forward re-end tail-end 'move)
1848 (setq tail-end (point))
1850 (when (and set-auto-coding-for-load
1851 (re-search-forward re-unibyte tail-end t))
1852 (setq coding-system 'raw-text))
1853 (when (and (not coding-system)
1854 (re-search-forward re-coding tail-end t))
1855 (setq coding-system (intern (match-string 1))))
1856 (when (and (not char-trans)
1857 (re-search-forward re-char-trans tail-end t))
1858 (setq char-trans (match-string 1))))))
1860 ;; If the coding-system name ends with "!", remove it and
1861 ;; set char-trans to "nil".
1862 (let ((name (symbol-name coding-system)))
1863 (if (= (aref name (1- (length name))) ?!)
1864 (setq coding-system (intern (substring name 0 -1))
1865 char-trans "nil"))))
1866 (when (and char-trans
1867 (not (setq char-trans (intern char-trans))))
1868 (make-local-variable 'enable-character-translation)
1869 (setq enable-character-translation nil))
1871 (cons coding-system :coding)))
1872 ;; Finally, try all the `auto-coding-functions'.
1873 (let ((funcs auto-coding-functions)
1874 (coding-system nil))
1875 (while (and funcs (not coding-system))
1876 (setq coding-system (condition-case e
1878 (goto-char (point-min))
1879 (funcall (pop funcs) size))
1882 (cons coding-system 'auto-coding-functions)))))
1884 (defun set-auto-coding (filename size)
1885 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1886 See `find-auto-coding' for how the coding system is found.
1887 Return nil if an invalid coding system is found.
1889 The variable `set-auto-coding-function' (which see) is set to this
1890 function by default."
1891 (let ((found (find-auto-coding filename size)))
1892 (if (and found (coding-system-p (car found)))
1895 (setq set-auto-coding-function 'set-auto-coding)
1897 (defun after-insert-file-set-coding (inserted &optional visit)
1898 "Set `buffer-file-coding-system' of current buffer after text is inserted.
1899 INSERTED is the number of characters that were inserted, as figured
1900 in the situation before this function. Return the number of characters
1901 inserted, as figured in the situation after. The two numbers can be
1902 different if the buffer has become unibyte.
1903 The optional second arg VISIT non-nil means that we are visiting a file."
1905 coding-system-for-read
1906 (not (eq coding-system-for-read 'auto-save-coding)))
1907 (setq buffer-file-coding-system-explicit
1908 (cons coding-system-for-read nil)))
1909 (if last-coding-system-used
1910 (let ((coding-system
1911 (find-new-buffer-file-coding-system last-coding-system-used)))
1913 (setq buffer-file-coding-system coding-system))))
1916 ;; The coding-spec and eol-type of coding-system returned is decided
1917 ;; independently in the following order.
1918 ;; 1. That of buffer-file-coding-system locally bound.
1919 ;; 2. That of CODING.
1921 (defun find-new-buffer-file-coding-system (coding)
1922 "Return a coding system for a buffer when a file of CODING is inserted.
1923 The local variable `buffer-file-coding-system' of the current buffer
1924 is set to the returned value.
1925 Return nil if there's no need to set `buffer-file-coding-system'."
1926 (let (local-coding local-eol
1927 found-coding found-eol
1930 ;; Nothing found about coding.
1933 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1934 ;; and LOCAL-CODING.
1935 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1936 (if (null (numberp local-eol))
1937 ;; But eol-type is not yet set.
1938 (setq local-eol nil))
1939 (if (and buffer-file-coding-system
1940 (not (eq (coding-system-type buffer-file-coding-system)
1942 (setq local-coding (coding-system-base buffer-file-coding-system)))
1944 (if (and (local-variable-p 'buffer-file-coding-system)
1945 local-eol local-coding)
1946 ;; The current buffer has already set full coding-system, we
1947 ;; had better not change it.
1950 (setq found-eol (coding-system-eol-type coding))
1951 (if (null (numberp found-eol))
1952 ;; But eol-type is not found.
1953 ;; If EOL conversions are inhibited, force unix eol-type.
1954 (setq found-eol (if inhibit-eol-conversion 0)))
1955 (setq found-coding (coding-system-base coding))
1957 (if (and (not found-eol) (eq found-coding 'undecided))
1958 ;; No valid coding information found.
1961 ;; Some coding information (eol or text) found.
1963 ;; The local setting takes precedence over the found one.
1964 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
1965 (or local-coding found-coding)
1966 (or found-coding local-coding)))
1967 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
1968 (or local-eol found-eol)
1969 (or found-eol local-eol)))
1971 (let ((eol-type (coding-system-eol-type new-coding)))
1972 (if (and (numberp new-eol) (vectorp eol-type))
1973 (aref eol-type new-eol)
1976 (defun modify-coding-system-alist (target-type regexp coding-system)
1977 "Modify one of look up tables for finding a coding system on I/O operation.
1978 There are three of such tables, `file-coding-system-alist',
1979 `process-coding-system-alist', and `network-coding-system-alist'.
1981 TARGET-TYPE specifies which of them to modify.
1982 If it is `file', it affects `file-coding-system-alist' (which see).
1983 If it is `process', it affects `process-coding-system-alist' (which see).
1984 If it is `network', it affects `network-coding-system-alist' (which see).
1986 REGEXP is a regular expression matching a target of I/O operation.
1987 The target is a file name if TARGET-TYPE is `file', a program name if
1988 TARGET-TYPE is `process', or a network service name or a port number
1989 to connect to if TARGET-TYPE is `network'.
1991 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1992 operation, or a cons cell (DECODING . ENCODING) specifying the coding
1993 systems for decoding and encoding respectively, or a function symbol
1994 which, when called, returns such a cons cell."
1995 (or (memq target-type '(file process network))
1996 (error "Invalid target type: %s" target-type))
1997 (or (stringp regexp)
1998 (and (eq target-type 'network) (integerp regexp))
1999 (error "Invalid regular expression: %s" regexp))
2000 (if (symbolp coding-system)
2001 (if (not (fboundp coding-system))
2003 (check-coding-system coding-system)
2004 (setq coding-system (cons coding-system coding-system))))
2005 (check-coding-system (car coding-system))
2006 (check-coding-system (cdr coding-system)))
2007 (cond ((eq target-type 'file)
2008 (let ((slot (assoc regexp file-coding-system-alist)))
2010 (setcdr slot coding-system)
2011 (setq file-coding-system-alist
2012 (cons (cons regexp coding-system)
2013 file-coding-system-alist)))))
2014 ((eq target-type 'process)
2015 (let ((slot (assoc regexp process-coding-system-alist)))
2017 (setcdr slot coding-system)
2018 (setq process-coding-system-alist
2019 (cons (cons regexp coding-system)
2020 process-coding-system-alist)))))
2022 (let ((slot (assoc regexp network-coding-system-alist)))
2024 (setcdr slot coding-system)
2025 (setq network-coding-system-alist
2026 (cons (cons regexp coding-system)
2027 network-coding-system-alist)))))))
2029 (defun decode-coding-inserted-region (from to filename
2030 &optional visit beg end replace)
2031 "Decode the region between FROM and TO as if it is read from file FILENAME.
2032 The idea is that the text between FROM and TO was just inserted somehow.
2033 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
2034 of the function `insert-file-contents'.
2035 Part of the job of this function is setting `buffer-undo-list' appropriately."
2038 (let ((coding coding-system-for-read)
2041 ;; Temporarily turn off undo recording, if we're decoding the
2042 ;; text of a visited file.
2043 (setq buffer-undo-list t)
2044 ;; Otherwise, if we can recognize the undo elt for the insertion,
2045 ;; remove it and get ready to replace it later.
2046 ;; In the mean time, turn off undo recording.
2047 (let ((last (car-safe buffer-undo-list)))
2048 (if (and (consp last) (eql (car last) from) (eql (cdr last) to))
2049 (setq undo-list-saved (cdr buffer-undo-list)
2050 buffer-undo-list t))))
2051 (narrow-to-region from to)
2052 (goto-char (point-min))
2054 (setq coding (funcall set-auto-coding-function
2055 filename (- (point-max) (point-min)))))
2057 (setq coding (car (find-operation-coding-system
2058 'insert-file-contents
2059 (cons filename (current-buffer))
2060 visit beg end replace))))
2061 (if (coding-system-p coding)
2062 (or enable-multibyte-characters
2064 (coding-system-change-text-conversion coding 'raw-text)))
2067 (decode-coding-region (point-min) (point-max) coding)
2068 (setq last-coding-system-used coding))
2069 ;; If we're decoding the text of a visited file,
2070 ;; the undo list should start out empty.
2072 (setq buffer-undo-list nil)
2073 ;; If we decided to replace the undo entry for the insertion,
2076 (setq buffer-undo-list
2077 (cons (cons from (point-max)) undo-list-saved))))))))
2079 (defun recode-region (start end new-coding coding)
2080 "Re-decode the region (previously decoded by CODING) by NEW-CODING."
2082 (list (region-beginning) (region-end)
2083 (read-coding-system "Text was really in: ")
2084 (let ((coding (or buffer-file-coding-system last-coding-system-used)))
2086 (concat "But was interpreted as"
2087 (if coding (format " (default %S): " coding) ": "))
2089 (or (and new-coding coding)
2090 (error "Coding system not specified"))
2091 ;; Check it before we encode the region.
2092 (check-coding-system new-coding)
2094 (narrow-to-region start end)
2095 (encode-coding-region (point-min) (point-max) coding)
2096 (decode-coding-region (point-min) (point-max) new-coding))
2097 (if (region-active-p)
2100 (defun make-translation-table (&rest args)
2101 "Make a translation table from arguments.
2102 A translation table is a char table intended for character
2103 translation in CCL programs.
2105 Each argument is a list of elements of the form (FROM . TO), where FROM
2106 is a character to be translated to TO.
2108 The arguments and forms in each argument are processed in the given
2109 order, and if a previous form already translates TO to some other
2110 character, say TO-ALT, FROM is also translated to TO-ALT."
2111 (let ((table (make-char-table 'translation-table))
2115 (let ((from (car elt))
2117 to-alt rev-from rev-to)
2118 ;; If we have already translated TO to TO-ALT, FROM should
2119 ;; also be translated to TO-ALT.
2120 (if (setq to-alt (aref table to))
2122 (aset table from to)
2123 ;; If we have already translated some chars to FROM, they
2124 ;; should also be translated to TO.
2125 (when (setq rev-from (assq from revlist))
2126 (dolist (elt (cdr rev-from))
2127 (aset table elt to))
2128 (setq revlist (delq rev-from revlist)
2129 rev-from (cdr rev-from)))
2130 ;; Now update REVLIST.
2131 (setq rev-to (assq to revlist))
2133 (setcdr rev-to (cons from (cdr rev-to)))
2134 (setq rev-to (list to from)
2135 revlist (cons rev-to revlist)))
2137 (setcdr rev-to (append rev-from (cdr rev-to)))))))
2138 ;; Return TABLE just created.
2139 (set-char-table-extra-slot table 1 1)
2142 (defun make-translation-table-from-vector (vec)
2143 "Make translation table from decoding vector VEC.
2144 VEC is an array of 256 elements to map unibyte codes to multibyte
2145 characters. Elements may be nil for undefined code points.
2146 See also the variable `nonascii-translation-table'."
2147 (let ((table (make-char-table 'translation-table))
2148 (rev-table (make-char-table 'translation-table))
2151 (setq ch (aref vec i))
2155 (aset rev-table ch i))))
2156 (set-char-table-extra-slot table 0 rev-table)
2157 (set-char-table-extra-slot table 1 1)
2158 (set-char-table-extra-slot rev-table 1 1)
2161 (defun make-translation-table-from-alist (alist)
2162 "Make translation table from N<->M mapping in ALIST.
2163 ALIST is an alist, each element has the form (FROM . TO).
2164 FROM and TO are a character or a vector of characters.
2165 If FROM is a character, that character is translated to TO.
2166 If FROM is a vector of characters, that sequence is translated to TO.
2167 The first extra-slot of the value is a translation table for reverse mapping."
2168 (let ((tables (vector (make-char-table 'translation-table)
2169 (make-char-table 'translation-table)))
2170 table max-lookup from to idx val)
2172 (setq table (aref tables i))
2176 (setq from (car elt) to (cdr elt))
2177 (setq from (cdr elt) to (car elt)))
2178 (if (characterp from)
2180 (setq idx (aref from 0)
2181 max-lookup (max max-lookup (length from))))
2182 (setq val (aref table idx))
2186 (setq val (list (cons (vector idx) val))))
2187 (if (characterp from)
2188 (setq from (vector from)))
2189 (setq val (nconc val (list (cons from to)))))
2190 (if (characterp from)
2192 (setq val (list (cons from to)))))
2193 (aset table idx val))
2194 (set-char-table-extra-slot table 1 max-lookup))
2195 (set-char-table-extra-slot (aref tables 0) 0 (aref tables 1))
2198 (defun define-translation-table (symbol &rest args)
2199 "Define SYMBOL as the name of translation table made by ARGS.
2200 This sets up information so that the table can be used for
2201 translations in a CCL program.
2203 If the first element of ARGS is a char-table whose purpose is
2204 `translation-table', just define SYMBOL to name it. (Note that this
2205 function does not bind SYMBOL.)
2207 Any other ARGS should be suitable as arguments of the function
2208 `make-translation-table' (which see).
2210 This function sets properties `translation-table' and
2211 `translation-table-id' of SYMBOL to the created table itself and the
2212 identification number of the table respectively. It also registers
2213 the table in `translation-table-vector'."
2214 (let ((table (if (and (char-table-p (car args))
2215 (eq (char-table-subtype (car args))
2216 'translation-table))
2218 (apply 'make-translation-table args)))
2219 (len (length translation-table-vector))
2222 (put symbol 'translation-table table)
2225 (setq translation-table-vector
2226 (vconcat translation-table-vector (make-vector len nil))))
2227 (let ((slot (aref translation-table-vector id)))
2229 (eq (car slot) symbol))
2231 (aset translation-table-vector id (cons symbol table))
2233 (setq id (1+ id)))))
2234 (put symbol 'translation-table-id id)
2237 (defun translate-region (start end table)
2238 "From START to END, translate characters according to TABLE.
2239 TABLE is a string or a char-table.
2240 If TABLE is a string, the Nth character in it is the mapping
2241 for the character with code N.
2242 If TABLE is a char-table, the element for character N is the mapping
2243 for the character with code N.
2244 It returns the number of characters changed."
2246 (list (region-beginning)
2249 (dotimes (i (length translation-table-vector))
2250 (if (consp (aref translation-table-vector i))
2251 (push (list (symbol-name
2252 (car (aref translation-table-vector i)))) l)))
2254 (error "No translation table defined"))
2256 (setq table (completing-read "Translation table: " l nil t)))
2259 (let ((val (get table 'translation-table)))
2260 (or (char-table-p val)
2261 (error "Invalid translation table name: %s" table))
2263 (translate-region-internal start end table))
2265 (put 'with-category-table 'lisp-indent-function 1)
2267 (defmacro with-category-table (table &rest body)
2268 "Execute BODY like `progn' with TABLE the current category table.
2269 The category table of the current buffer is saved, BODY is evaluated,
2270 then the saved table is restored, even in case of an abnormal exit.
2271 Value is what BODY returns."
2272 (let ((old-table (make-symbol "old-table"))
2273 (old-buffer (make-symbol "old-buffer")))
2274 `(let ((,old-table (category-table))
2275 (,old-buffer (current-buffer)))
2278 (set-category-table ,table)
2280 (with-current-buffer ,old-buffer
2281 (set-category-table ,old-table))))))
2283 (defun define-translation-hash-table (symbol table)
2284 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2286 Analogous to `define-translation-table', but updates
2287 `translation-hash-table-vector' and the table is for use in the CCL
2288 `lookup-integer' and `lookup-character' functions."
2289 (unless (and (symbolp symbol)
2290 (hash-table-p table))
2291 (error "Bad args to define-translation-hash-table"))
2292 (let ((len (length translation-hash-table-vector))
2295 (put symbol 'translation-hash-table table)
2298 (setq translation-hash-table-vector
2299 (vconcat translation-hash-table-vector [nil])))
2300 (let ((slot (aref translation-hash-table-vector id)))
2302 (eq (car slot) symbol))
2304 (aset translation-hash-table-vector id (cons symbol table))
2306 (setq id (1+ id)))))
2307 (put symbol 'translation-hash-table-id id)
2310 ;;; Initialize some variables.
2312 (put 'use-default-ascent 'char-table-extra-slots 0)
2313 (setq use-default-ascent (make-char-table 'use-default-ascent))
2314 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2315 (setq ignore-relative-composition
2316 (make-char-table 'ignore-relative-composition))
2318 (make-obsolete 'set-char-table-default
2319 "generic characters no longer exist." "23.1")
2321 ;;; Built-in auto-coding-functions:
2323 (defun sgml-xml-auto-coding-function (size)
2324 "Determine whether the buffer is XML, and if so, its encoding.
2325 This function is intended to be added to `auto-coding-functions'."
2326 (setq size (+ (point) size))
2327 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" size t)
2328 (let ((end (save-excursion
2330 (re-search-forward "[\"']\\s-*\\?>" size t))))
2332 (if (re-search-forward "encoding=[\"']\\(.+?\\)[\"']" end t)
2333 (let* ((match (match-string 1))
2334 (sym (intern (downcase match))))
2335 (if (coding-system-p sym)
2337 (message "Warning: unknown coding system \"%s\"" match)
2339 ;; Files without an encoding tag should be UTF-8. But users
2340 ;; may be naive about encodings, and have saved the file from
2341 ;; another editor that does not help them get the encoding right.
2342 ;; Detect the encoding and warn the user if it is detected as
2343 ;; something other than UTF-8.
2345 (with-coding-priority '(utf-8)
2347 (detect-coding-region (point-min) size t)))))
2348 ;; Pure ASCII always comes back as undecided.
2349 (if (memq detected '(utf-8 undecided))
2351 (warn "File contents detected as %s.
2352 Consider adding an encoding attribute to the xml declaration,
2353 or saving as utf-8, as mandated by the xml specification." detected)
2356 (defun sgml-html-meta-auto-coding-function (size)
2357 "If the buffer has an HTML meta tag, use it to determine encoding.
2358 This function is intended to be added to `auto-coding-functions'."
2359 (let ((case-fold-search t))
2360 (setq size (min (+ (point) size)
2362 ;; Limit the search by the end of the HTML header.
2363 (or (search-forward "</head>" (+ (point) size) t)
2364 ;; In case of no header, search only 10 lines.
2367 ;; Make sure that the buffer really contains an HTML document, by
2368 ;; checking that it starts with a doctype or a <HTML> start tag
2369 ;; (allowing for whitespace at bob). Note: 'DOCTYPE NETSCAPE' is
2370 ;; useful for Mozilla bookmark files.
2371 (when (and (re-search-forward "\\`[[:space:]\n]*\\(<!doctype[[:space:]\n]+\\(html\\|netscape\\)\\|<html\\)" size t)
2372 (re-search-forward "<meta\\s-+http-equiv=[\"']?content-type[\"']?\\s-+content=[\"']text/\\sw+;\\s-*charset=\\(.+?\\)[\"']" size t))
2373 (let* ((match (match-string 1))
2374 (sym (intern (downcase match))))
2375 (if (coding-system-p sym)
2377 (message "Warning: unknown coding system \"%s\"" match)
2380 (defun xml-find-file-coding-system (args)
2381 "Determine the coding system of an XML file without a declaration.
2382 Strictly speaking, the file should be utf-8, but mistakes are
2383 made, and there are genuine cases where XML fragments are saved,
2384 with the encoding properly specified in a master document, or
2385 added by processing software."
2386 (if (eq (car args) 'insert-file-contents)
2388 (with-coding-priority '(utf-8)
2390 (detect-coding-region (point-min) (point-max) t)))))
2391 ;; Pure ASCII always comes back as undecided.
2393 ((memq detected '(utf-8 undecided))
2395 ((eq detected 'utf-16le-with-signature) 'utf-16le-with-signature)
2396 ((eq detected 'utf-16be-with-signature) 'utf-16be-with-signature)
2398 (warn "File contents detected as %s.
2399 Consider adding an xml declaration with the encoding specified,
2400 or saving as utf-8, as mandated by the xml specification." detected)
2402 ;; Don't interfere with the user's wishes for saving the buffer.
2403 ;; We did what we could when the buffer was created to ensure the
2404 ;; correct encoding was used, or the user was warned, so any
2405 ;; non-conformity here is deliberate on the part of the user.
2411 ;; arch-tag: 9aebaa6e-0e8a-40a9-b857-cb5d04a39e7c
2412 ;;; mule.el ends here