1 ;;; mule.el --- basic commands for multilingual environment
3 ;; Copyright (C) 1997-2013 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
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/>.
33 ;; FIXME? Are these still relevant? Nothing uses them AFAICS.
34 (defconst mule-version
"6.0 (HANACHIRUSATO)" "\
35 Version number and name of this version of MULE (multilingual environment).")
37 (defconst mule-version-date
"2003.9.1" "\
38 Distribution date of this version of MULE (multilingual environment).")
43 ;; Backward compatibility code for handling emacs-mule charsets.
44 (defvar private-char-area-1-min
#xF0000
)
45 (defvar private-char-area-1-max
#xFFFFE
)
46 (defvar private-char-area-2-min
#x100000
)
47 (defvar private-char-area-2-max
#x10FFFE
)
49 ;; Table of emacs-mule charsets indexed by their emacs-mule ID.
50 (defvar emacs-mule-charset-table
(make-vector 256 nil
))
51 (aset emacs-mule-charset-table
0 'ascii
)
53 ;; Convert the argument of old-style call of define-charset to a
54 ;; property list used by the new-style.
55 ;; INFO-VECTOR is a vector of the format:
56 ;; [DIMENSION CHARS WIDTH DIRECTION ISO-FINAL-CHAR ISO-GRAPHIC-PLANE
57 ;; SHORT-NAME LONG-NAME DESCRIPTION]
59 (defun convert-define-charset-argument (emacs-mule-id info-vector
)
60 (let* ((dim (aref info-vector
0))
61 (chars (aref info-vector
1))
62 (total (if (= dim
1) chars
(* chars chars
)))
63 (code-space (if (= dim
1) (if (= chars
96) [32 127] [33 126])
64 (if (= chars
96) [32 127 32 127] [33 126 33 126])))
66 (if (integerp emacs-mule-id
)
67 (or (= emacs-mule-id
0)
68 (and (>= emacs-mule-id
129) (< emacs-mule-id
256))
69 (error "Invalid CHARSET-ID: %d" emacs-mule-id
))
71 (if (= dim
1) (setq from-id
160 to-id
224)
72 (setq from-id
224 to-id
255))
73 (while (and (< from-id to-id
)
74 (not (aref emacs-mule-charset-table from-id
)))
75 (setq from-id
(1+ from-id
)))
77 (error "No more room for the new Emacs-mule charset"))
78 (setq emacs-mule-id from-id
)))
79 (if (> (- private-char-area-1-max private-char-area-1-min
) total
)
80 (setq code-offset private-char-area-1-min
81 private-char-area-1-min
(+ private-char-area-1-min total
))
82 (if (> (- private-char-area-2-max private-char-area-2-min
) total
)
83 (setq code-offset private-char-area-2-min
84 private-char-area-2-min
(+ private-char-area-2-min total
))
85 (error "No more space for a new charset")))
87 :code-space code-space
88 :iso-final-char
(aref info-vector
4)
89 :code-offset code-offset
90 :emacs-mule-id emacs-mule-id
)))
92 (defun define-charset (name docstring
&rest props
)
93 "Define NAME (symbol) as a charset with DOCSTRING.
94 The remaining arguments must come in pairs ATTRIBUTE VALUE. ATTRIBUTE
95 may be any symbol. The following have special meanings, and one of
96 `:code-offset', `:map', `:subset', `:superset' must be specified.
100 VALUE must be a short string to identify the charset. If omitted,
105 VALUE must be a string longer than `:short-name' to identify the
106 charset. If omitted, the value of the `:short-name' attribute is used.
110 VALUE must be an integer 0, 1, 2, or 3, specifying the dimension of
111 code-points of the charsets. If omitted, it is calculated from the
112 value of the `:code-space' attribute.
116 VALUE must be a vector of length at most 8 specifying the byte code
117 range of each dimension in this format:
118 [ MIN-1 MAX-1 MIN-2 MAX-2 ... ]
119 where MIN-N is the minimum byte value of Nth dimension of code-point,
120 MAX-N is the maximum byte value of that.
124 VALUE must be an integer specifying the minimum code point of the
125 charset. If omitted, it is calculated from `:code-space'. VALUE may
126 be a cons (HIGH . LOW), where HIGH is the most significant 16 bits of
127 the code point and LOW is the least significant 16 bits.
131 VALUE must be an integer specifying the maximum code point of the
132 charset. If omitted, it is calculated from `:code-space'. VALUE may
133 be a cons (HIGH . LOW), where HIGH is the most significant 16 bits of
134 the code point and LOW is the least significant 16 bits.
138 VALUE must be a character in the range 32 to 127 (inclusive)
139 specifying the final char of the charset for ISO-2022 encoding. If
140 omitted, the charset can't be encoded by ISO-2022 based
143 `:iso-revision-number'
145 VALUE must be an integer in the range 0..63, specifying the revision
146 number of the charset for ISO-2022 encoding.
150 VALUE must be an integer of 0, 129..255. If omitted, the charset
151 can't be encoded by coding-systems of type `emacs-mule'.
153 `:ascii-compatible-p'
155 VALUE must be nil or t (default nil). If VALUE is t, the charset is
156 compatible with ASCII, i.e. the first 128 code points map to ASCII.
160 VALUE must be nil or t. If the VALUE is t, the charset is
161 supplementary, which means it is used only as a parent or a
162 subset of some other charset, or it is provided just for backward
167 VALUE must be a nonnegative integer that can be used as an invalid
168 code point of the charset. If the minimum code is 0 and the maximum
169 code is greater than Emacs's maximum integer value, `:invalid-code'
170 should not be omitted.
174 VALUE must be an integer added to the index number of a character to
175 get the corresponding character code.
179 VALUE must be vector or string.
181 If it is a vector, the format is [ CODE-1 CHAR-1 CODE-2 CHAR-2 ... ],
182 where CODE-n is a code-point of the charset, and CHAR-n is the
183 corresponding character code.
185 If it is a string, it is a name of file that contains the above
186 information. Each line of the file must be this format:
188 where XXX is a hexadecimal representation of CODE-n and YYY is a
189 hexadecimal representation of CHAR-n. A line starting with `#' is a
194 VALUE must be a list:
195 ( PARENT MIN-CODE MAX-CODE OFFSET )
196 PARENT is a parent charset. MIN-CODE and MAX-CODE specify the range
197 of characters inherited from the parent. OFFSET is an integer value
198 to add to a code point of the parent charset to get the corresponding
199 code point of this charset.
203 VALUE must be a list of parent charsets. The charset inherits
204 characters from them. Each element of the list may be a cons (PARENT
205 . OFFSET), where PARENT is a parent charset, and OFFSET is an offset
206 value to add to a code point of PARENT to get the corresponding code
207 point of this charset.
211 VALUE must be vector or string.
213 If it is a vector, the format is [ CODE-1 CHAR-1 CODE-2 CHAR-2 ... ],
214 where CODE-n is a code-point of the charset, and CHAR-n is the
215 corresponding Unicode character code.
217 If it is a string, it is a name of file that contains the above
218 information. The file format is the same as what described for `:map'
220 (when (vectorp (car props
))
222 ;; (define-charset CHARSET-ID CHARSET-SYMBOL INFO-VECTOR)
223 ;; Convert the argument to make it fit with the current style.
224 (let ((vec (car props
)))
225 (setq props
(convert-define-charset-argument name vec
)
227 docstring
(aref vec
8))))
228 (let ((attrs (mapcar 'list
'(:dimension
245 ;; If :dimension is omitted, get the dimension from :code-space.
246 (let ((dimension (plist-get props
:dimension
)))
248 (let ((code-space (plist-get props
:code-space
)))
249 (setq dimension
(if code-space
(/ (length code-space
) 2) 4))
250 (setq props
(plist-put props
:dimension dimension
)))))
252 (let ((code-space (plist-get props
:code-space
)))
254 (let ((dimension (plist-get props
:dimension
)))
255 (setq code-space
(make-vector 8 0))
256 (dotimes (i dimension
)
257 (aset code-space
(1+ (* i
2)) #xFF
))
258 (setq props
(plist-put props
:code-space code-space
)))))
260 ;; If :emacs-mule-id is specified, update emacs-mule-charset-table.
261 (let ((emacs-mule-id (plist-get props
:emacs-mule-id
)))
262 (if (integerp emacs-mule-id
)
263 (aset emacs-mule-charset-table emacs-mule-id name
)))
266 (setcdr slot
(purecopy (plist-get props
(car slot
)))))
268 ;; Make sure that the value of :code-space is a vector of 8
270 (let* ((slot (assq :code-space attrs
))
275 (vconcat val
(make-vector (- 8 len
) 0)))))
277 ;; Add :name and :docstring properties to PROPS.
279 (cons :name
(cons name
(cons :docstring
(cons (purecopy docstring
) props
)))))
280 (or (plist-get props
:short-name
)
281 (plist-put props
:short-name
(symbol-name name
)))
282 (or (plist-get props
:long-name
)
283 (plist-put props
:long-name
(plist-get props
:short-name
)))
284 (plist-put props
:base 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 (set-buffer-multibyte t
)
330 ;; Don't let deactivate-mark remain set.
331 (let (deactivate-mark)
332 (insert-file-contents fullname
))
333 ;; If the loaded file was inserted with no-conversion or
334 ;; raw-text coding system, make the buffer unibyte.
335 ;; Otherwise, eval-buffer might try to interpret random
336 ;; binary junk as multibyte characters.
337 (if (and enable-multibyte-characters
338 (or (eq (coding-system-type last-coding-system-used
)
340 (set-buffer-multibyte nil
))
341 ;; Make `kill-buffer' quiet.
342 (set-buffer-modified-p nil
))
343 ;; Have the original buffer current while we eval.
344 (eval-buffer buffer nil
345 ;; This is compatible with what `load' does.
346 (if purify-flag file fullname
)
348 (let (kill-buffer-hook kill-buffer-query-functions
)
349 (kill-buffer buffer
)))
350 (do-after-load-evaluation fullname
)
352 (unless (or nomessage noninteractive
)
354 (message "Loading %s (source)...done" file
)
355 (message "Loading %s...done" file
)))
358 (defun charset-info (charset)
359 "Return a vector of information of CHARSET.
360 This function is provided for backward compatibility.
362 The elements of the vector are:
363 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
364 LEADING-CODE-BASE, LEADING-CODE-EXT,
365 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
366 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
369 CHARSET-ID is always 0.
371 DIMENSION is the number of bytes of a code-point of the charset:
373 CHARS is the number of characters in a dimension:
376 DIRECTION is always 0.
377 LEADING-CODE-BASE is always 0.
378 LEADING-CODE-EXT is always 0.
379 ISO-FINAL-CHAR (character) is the final character of the
380 corresponding ISO 2022 charset. If the charset is not assigned
381 any final character, the value is -1.
382 ISO-GRAPHIC-PLANE is always 0.
383 REVERSE-CHARSET is always -1.
384 SHORT-NAME (string) is the short name to refer to the charset.
385 LONG-NAME (string) is the long name to refer to the charset
386 DESCRIPTION (string) is the description string of the charset.
387 PLIST (property list) may contain any type of information a user
388 want to put and get by functions `put-charset-property' and
389 `get-charset-property' respectively."
392 (charset-dimension charset
)
393 (charset-chars charset
)
398 (charset-iso-final-char charset
)
401 (get-charset-property charset
:short-name
)
402 (get-charset-property charset
:short-name
)
403 (charset-description charset
)
404 (charset-plist charset
)))
406 ;; It is better not to use backquote in this file,
407 ;; because that makes a bootstrapping problem
408 ;; if you need to recompile all the Lisp files using interpreted code.
410 (defun charset-id (charset)
411 "Always return 0. This is provided for backward compatibility."
412 (declare (obsolete nil
"23.1"))
415 (defmacro charset-bytes
(charset)
416 "Always return 0. This is provided for backward compatibility."
417 (declare (obsolete nil
"23.1"))
420 (defun get-charset-property (charset propname
)
421 "Return the value of CHARSET's PROPNAME property.
422 This is the last value stored with
423 (put-charset-property CHARSET PROPNAME VALUE)."
424 (plist-get (charset-plist charset
) propname
))
426 (defun put-charset-property (charset propname value
)
427 "Set CHARSETS's PROPNAME property to value VALUE.
428 It can be retrieved with `(get-charset-property CHARSET PROPNAME)'."
429 (set-charset-plist charset
430 (plist-put (charset-plist charset
) propname
435 (defun charset-description (charset)
436 "Return description string of CHARSET."
437 (plist-get (charset-plist charset
) :docstring
))
439 (defun charset-dimension (charset)
440 "Return dimension of CHARSET."
441 (plist-get (charset-plist charset
) :dimension
))
443 (defun charset-chars (charset &optional dimension
)
444 "Return number of characters contained in DIMENSION of CHARSET.
445 DIMENSION defaults to the first dimension."
446 (unless dimension
(setq dimension
1))
447 (let ((code-space (plist-get (charset-plist charset
) :code-space
)))
448 (1+ (- (aref code-space
(1- (* 2 dimension
)))
449 (aref code-space
(- (* 2 dimension
) 2))))))
451 (defun charset-iso-final-char (charset)
452 "Return ISO-2022 final character of CHARSET.
453 Return -1 if charset isn't an ISO 2022 one."
454 (or (plist-get (charset-plist charset
) :iso-final-char
)
457 (defmacro charset-short-name
(charset)
458 "Return short name of CHARSET."
459 (plist-get (charset-plist charset
) :short-name
))
461 (defmacro charset-long-name
(charset)
462 "Return long name of CHARSET."
463 (plist-get (charset-plist charset
) :long-name
))
465 (defun charset-list ()
466 "Return list of all charsets ever defined."
467 (declare (obsolete charset-list
"23.1"))
472 (define-obsolete-function-alias 'char-valid-p
'characterp
"23.1")
474 (defun generic-char-p (char)
475 "Always return nil. This is provided for backward compatibility."
476 (declare (obsolete nil
"23.1"))
479 (defun make-char-internal (charset-id &optional code1 code2
)
480 (let ((charset (aref emacs-mule-charset-table charset-id
)))
482 (error "Invalid Emacs-mule charset ID: %d" charset-id
))
483 (make-char charset code1 code2
)))
485 ;; Save the ASCII case table in case we need it later. Some locales
486 ;; (such as Turkish) modify the case behavior of ASCII characters,
487 ;; which can interfere with networking code that uses ASCII strings.
489 (defvar ascii-case-table
490 ;; Code copied from copy-case-table to avoid requiring case-table.el
491 (let ((tbl (copy-sequence (standard-case-table)))
492 (up (char-table-extra-slot (standard-case-table) 0)))
493 (if up
(set-char-table-extra-slot tbl
0 (copy-sequence up
)))
494 (set-char-table-extra-slot tbl
1 nil
)
495 (set-char-table-extra-slot tbl
2 nil
)
497 "Case table for the ASCII character set.")
499 ;; Coding system stuff
501 ;; Coding system is a symbol that has been defined by the function
502 ;; `define-coding-system'.
504 (defconst coding-system-iso-2022-flags
522 "List of symbols that control ISO-2022 encoder/decoder.
524 The value of the `:flags' attribute in the argument of the function
525 `define-coding-system' must be one of them.
527 If `long-form' is specified, use a long designation sequence on
528 encoding for the charsets `japanese-jisx0208-1978', `chinese-gb2312',
529 and `japanese-jisx0208'. The long designation sequence doesn't
530 conform to ISO 2022, but is used by such coding systems as
533 If `ascii-at-eol' is specified, designate ASCII to g0 at end of line
536 If `ascii-at-cntl' is specified, designate ASCII to g0 before control
537 codes and SPC on encoding.
539 If `7-bit' is specified, use 7-bit code only on encoding.
541 If `locking-shift' is specified, decode locking-shift code correctly
542 on decoding, and use locking-shift to invoke a graphic element on
545 If `single-shift' is specified, decode single-shift code correctly on
546 decoding, and use single-shift to invoke a graphic element on encoding.
548 If `designation' is specified, decode designation code correctly on
549 decoding, and use designation to designate a charset to a graphic
552 If `revision' is specified, produce an escape sequence to specify
553 revision number of a charset on encoding. Such an escape sequence is
554 always correctly decoded on decoding.
556 If `direction' is specified, decode ISO6429's code for specifying
557 direction correctly, and produce the code on encoding.
559 If `init-at-bol' is specified, on encoding, it is assumed that
560 invocation and designation statuses are reset at each beginning of
561 line even if `ascii-at-eol' is not specified; thus no codes for
562 resetting them are produced.
564 If `safe' is specified, on encoding, characters not supported by a
565 coding are replaced with `?'.
567 If `latin-extra' is specified, the code-detection routine assumes that a
568 code specified in `latin-extra-code-table' (which see) is valid.
570 If `composition' is specified, an escape sequence to specify
571 composition sequence is correctly decoded on decoding, and is produced
574 If `euc-tw-shift' is specified, the EUC-TW specific shifting code is
575 correctly decoded on decoding, and is produced on encoding.
577 If `use-roman' is specified, JIS0201-1976-Roman is designated instead
580 If `use-oldjis' is specified, JIS0208-1976 is designated instead of
583 (defun define-coding-system (name docstring
&rest props
)
584 "Define NAME (a symbol) as a coding system with DOCSTRING and attributes.
585 The remaining arguments must come in pairs ATTRIBUTE VALUE. ATTRIBUTE
588 The following attributes have special meanings. Those labeled as
589 \"(required)\" should not be omitted.
591 `:mnemonic' (required)
593 VALUE is a character to display on mode line for the coding system.
595 `:coding-type' (required)
597 VALUE must be one of `charset', `utf-8', `utf-16', `iso-2022',
598 `emacs-mule', `shift-jis', `ccl', `raw-text', `undecided'.
602 VALUE is the EOL (end-of-line) format of the coding system. It must be
603 one of `unix', `dos', `mac'. The symbol `unix' means Unix-like EOL
604 \(i.e. single LF), `dos' means DOS-like EOL \(i.e. sequence of CR LF),
605 and `mac' means Mac-like EOL \(i.e. single CR). If omitted, Emacs
606 detects the EOL format automatically when decoding.
610 VALUE must be a list of charsets supported by the coding system. On
611 encoding by the coding system, if a character belongs to multiple
612 charsets in the list, a charset that comes earlier in the list is
613 selected. If `:coding-type' is `iso-2022', VALUE may be `iso-2022',
614 which indicates that the coding system supports all ISO-2022 based
615 charsets. If `:coding-type' is `emacs-mule', VALUE may be
616 `emacs-mule', which indicates that the coding system supports all
617 charsets that have the `:emacs-mule-id' property.
619 `:ascii-compatible-p'
621 If VALUE is non-nil, the coding system decodes all 7-bit bytes into
622 the corresponding ASCII characters, and encodes all ASCII characters
623 back to the corresponding 7-bit bytes. VALUE defaults to nil.
625 `:decode-translation-table'
627 VALUE must be a translation table to use on decoding.
629 `:encode-translation-table'
631 VALUE must be a translation table to use on encoding.
633 `:post-read-conversion'
635 VALUE must be a function to call after some text is inserted and
636 decoded by the coding system itself and before any functions in
637 `after-insert-functions' are called. This function is passed one
638 argument; the number of characters in the text to convert, with
639 point at the start of the text. The function should leave point
640 the same, and return the new character count.
642 `:pre-write-conversion'
644 VALUE must be a function to call after all functions in
645 `write-region-annotate-functions' and `buffer-file-format' are
646 called, and before the text is encoded by the coding system
647 itself. This function should convert the whole text in the
648 current buffer. For backward compatibility, this function is
649 passed two arguments which can be ignored.
653 VALUE must be a character. On encoding, a character not supported by
654 the coding system is replaced with VALUE.
658 VALUE non-nil means that visiting a file with the coding system
659 results in a unibyte buffer.
663 VALUE must be a symbol whose name is that of a MIME charset converted
666 `:mime-text-unsuitable'
668 VALUE non-nil means the `:mime-charset' property names a charset which
669 is unsuitable for the top-level media type \"text\".
673 VALUE must be a list of symbols that control the ISO-2022 converter.
674 Each must be a member of the list `coding-system-iso-2022-flags'
675 \(which see). This attribute has a meaning only when `:coding-type'
680 VALUE must be a vector [G0-USAGE G1-USAGE G2-USAGE G3-USAGE].
681 GN-USAGE specifies the usage of graphic register GN as follows.
683 If it is nil, no charset can be designated to GN.
685 If it is a charset, the charset is initially designated to GN, and
686 never used by the other charsets.
688 If it is a list, the elements must be charsets, nil, 94, or 96. GN
689 can be used by all the listed charsets. If the list contains 94, any
690 iso-2022 charset whose code-space ranges are 94 long can be designated
691 to GN. If the list contains 96, any charsets whose whose ranges are
692 96 long can be designated to GN. If the first element is a charset,
693 that charset is initially designated to GN.
695 This attribute has a meaning only when `:coding-type' is `iso-2022'.
699 This attributes specifies whether the coding system uses a `byte order
700 mark'. VALUE must be nil, t, or cons of coding systems whose
701 `:coding-type' is `utf-16' or `utf-8'.
703 If the value is nil, on decoding, don't treat the first two-byte as
704 BOM, and on encoding, don't produce BOM bytes.
706 If the value is t, on decoding, skip the first two-byte as BOM, and on
707 encoding, produce BOM bytes according to the value of `:endian'.
709 If the value is cons, on decoding, check the first two-byte. If they
710 are 0xFE 0xFF, use the car part coding system of the value. If they
711 are 0xFF 0xFE, use the cdr part coding system of the value.
712 Otherwise, treat them as bytes for a normal character. On encoding,
713 produce BOM bytes according to the value of `:endian'.
715 This attribute has a meaning only when `:coding-type' is `utf-16' or
720 VALUE must be `big' or `little' specifying big-endian and
721 little-endian respectively. The default value is `big'.
723 This attribute has a meaning only when `:coding-type' is `utf-16'.
727 VALUE is a symbol representing the registered CCL program used for
728 decoding. This attribute has a meaning only when `:coding-type' is
733 VALUE is a symbol representing the registered CCL program used for
734 encoding. This attribute has a meaning only when `:coding-type' is
736 (let* ((common-attrs (mapcar 'list
741 :decode-translation-table
742 :encode-translation-table
743 :post-read-conversion
744 :pre-write-conversion
749 (coding-type (plist-get props
:coding-type
))
750 (spec-attrs (mapcar 'list
751 (cond ((eq coding-type
'iso-2022
)
756 ((eq coding-type
'utf-8
)
758 ((eq coding-type
'utf-16
)
761 ((eq coding-type
'ccl
)
766 (dolist (slot common-attrs
)
767 (setcdr slot
(plist-get props
(car slot
))))
769 (dolist (slot spec-attrs
)
770 (setcdr slot
(plist-get props
(car slot
))))
772 (if (eq coding-type
'iso-2022
)
773 (let ((designation (plist-get props
:designation
))
774 (flags (plist-get props
:flags
))
775 (initial (make-vector 4 nil
))
776 (reg-usage (cons 4 4))
779 (setq elt
(aref designation i
))
780 (cond ((charsetp elt
)
782 (setq request
(cons (cons elt i
) request
)))
784 (aset initial i
(car elt
))
785 (if (charsetp (car elt
))
786 (setq request
(cons (cons (car elt
) i
) request
)))
787 (dolist (e (cdr elt
))
789 (setq request
(cons (cons e i
) request
)))
791 (setcar reg-usage i
))
793 (setcdr reg-usage i
))
796 (setcdr reg-usage i
)))))))
797 (setcdr (assq :initial spec-attrs
) initial
)
798 (setcdr (assq :reg-usage spec-attrs
) reg-usage
)
799 (setcdr (assq :request spec-attrs
) request
)
801 ;; Change :flags value from a list to a bit-mask.
804 (dolist (elt coding-system-iso-2022-flags
)
806 (setq bits
(logior bits
(lsh 1 i
))))
808 (setcdr (assq :flags spec-attrs
) bits
))))
810 ;; Add :name and :docstring properties to PROPS.
812 (cons :name
(cons name
(cons :docstring
(cons (purecopy docstring
)
814 (setcdr (assq :plist common-attrs
) props
)
815 (apply 'define-coding-system-internal
816 name
(mapcar 'cdr
(append common-attrs spec-attrs
)))))
818 (defun coding-system-doc-string (coding-system)
819 "Return the documentation string for CODING-SYSTEM."
820 (plist-get (coding-system-plist coding-system
) :docstring
))
822 (defun coding-system-mnemonic (coding-system)
823 "Return the mnemonic character of CODING-SYSTEM.
824 The mnemonic character of a coding system is used in mode line to
825 indicate the coding system. If CODING-SYSTEM is nil, return ?=."
826 (plist-get (coding-system-plist coding-system
) :mnemonic
))
828 (defun coding-system-type (coding-system)
829 "Return the coding type of CODING-SYSTEM.
830 A coding type is a symbol indicating the encoding method of CODING-SYSTEM.
831 See the function `define-coding-system' for more detail."
832 (plist-get (coding-system-plist coding-system
) :coding-type
))
834 (defun coding-system-charset-list (coding-system)
835 "Return list of charsets supported by CODING-SYSTEM.
836 If CODING-SYSTEM supports all ISO-2022 charsets, return `iso-2022'.
837 If CODING-SYSTEM supports all emacs-mule charsets, return `emacs-mule'."
838 (plist-get (coding-system-plist coding-system
) :charset-list
))
840 (defun coding-system-category (coding-system)
841 "Return a category symbol of CODING-SYSTEM."
842 (plist-get (coding-system-plist coding-system
) :category
))
844 (defun coding-system-get (coding-system prop
)
845 "Extract a value from CODING-SYSTEM's property list for property PROP.
846 For compatibility with Emacs 20/21, this accepts old-style symbols
847 like `mime-charset' as well as the current style like `:mime-charset'."
848 (or (plist-get (coding-system-plist coding-system
) prop
)
849 (if (not (keywordp prop
))
850 ;; For backward compatibility.
851 (if (eq prop
'ascii-incompatible
)
852 (not (plist-get (coding-system-plist coding-system
)
853 :ascii-compatible-p
))
854 (plist-get (coding-system-plist coding-system
)
855 (intern (concat ":" (symbol-name prop
))))))))
857 (defun coding-system-eol-type-mnemonic (coding-system)
858 "Return the string indicating end-of-line format of CODING-SYSTEM."
859 (let* ((eol-type (coding-system-eol-type coding-system
))
860 (val (cond ((eq eol-type
0) eol-mnemonic-unix
)
861 ((eq eol-type
1) eol-mnemonic-dos
)
862 ((eq eol-type
2) eol-mnemonic-mac
)
863 (t eol-mnemonic-undecided
))))
866 (char-to-string val
))))
868 (defun coding-system-lessp (x y
)
869 (cond ((eq x
'no-conversion
) t
)
870 ((eq y
'no-conversion
) nil
)
871 ((eq x
'emacs-mule
) t
)
872 ((eq y
'emacs-mule
) nil
)
873 ((eq x
'undecided
) t
)
874 ((eq y
'undecided
) nil
)
875 (t (let ((c1 (coding-system-mnemonic x
))
876 (c2 (coding-system-mnemonic y
)))
877 (or (< (downcase c1
) (downcase c2
))
878 (and (not (> (downcase c1
) (downcase c2
)))
881 (defun coding-system-equal (coding-system-1 coding-system-2
)
882 "Return t if and only if CODING-SYSTEM-1 and CODING-SYSTEM-2 are identical.
883 Two coding systems are identical if both symbols are equal
884 or one is an alias of the other."
885 (or (eq coding-system-1 coding-system-2
)
886 (and (equal (coding-system-plist coding-system-1
)
887 (coding-system-plist coding-system-2
))
888 (let ((eol-type-1 (coding-system-eol-type coding-system-1
))
889 (eol-type-2 (coding-system-eol-type coding-system-2
)))
890 (or (eq eol-type-1 eol-type-2
)
891 (and (vectorp eol-type-1
) (vectorp eol-type-2
)))))))
893 (defun add-to-coding-system-list (coding-system)
894 "Add CODING-SYSTEM to variable `coding-system-list' while keeping it sorted."
895 (if (or (null coding-system-list
)
896 (coding-system-lessp coding-system
(car coding-system-list
)))
897 (setq coding-system-list
(cons coding-system coding-system-list
))
898 (let ((len (length coding-system-list
))
899 mid
(tem coding-system-list
))
901 (setq mid
(nthcdr (/ len
2) tem
))
902 (if (coding-system-lessp (car mid
) coding-system
)
904 len
(- len
(/ len
2)))
905 (setq len
(/ len
2))))
906 (setcdr tem
(cons coding-system
(cdr tem
))))))
908 (defun coding-system-list (&optional base-only
)
909 "Return a list of all existing non-subsidiary coding systems.
910 If optional arg BASE-ONLY is non-nil, only base coding systems are
911 listed. The value doesn't include subsidiary coding systems which are
912 made from bases and aliases automatically for various end-of-line
913 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
915 (dolist (coding coding-system-list
)
916 (if (eq (coding-system-base coding
) coding
)
918 (setq codings
(cons coding codings
))
919 (dolist (alias (coding-system-aliases coding
))
920 (setq codings
(cons alias codings
))))))
923 (defconst char-coding-system-table nil
924 "It exists just for backward compatibility, and the value is always nil.")
925 (make-obsolete-variable 'char-coding-system-table nil
"23.1")
927 (defun transform-make-coding-system-args (name type
&optional doc-string props
)
928 "For internal use only.
929 Transform XEmacs style args for `make-coding-system' to Emacs style.
930 Value is a list of transformed arguments."
931 (let ((mnemonic (string-to-char (or (plist-get props
'mnemonic
) "?")))
932 (eol-type (plist-get props
'eol-type
))
935 ((eq eol-type
'lf
) (setq eol-type
'unix
))
936 ((eq eol-type
'crlf
) (setq eol-type
'dos
))
937 ((eq eol-type
'cr
) (setq eol-type
'mac
)))
938 (if (setq tmp
(plist-get props
'post-read-conversion
))
939 (setq properties
(plist-put properties
'post-read-conversion tmp
)))
940 (if (setq tmp
(plist-get props
'pre-write-conversion
))
941 (setq properties
(plist-put properties
'pre-write-conversion tmp
)))
943 ((eq type
'shift-jis
)
944 `(,name
1 ,mnemonic
,doc-string
() ,properties
,eol-type
))
945 ((eq type
'iso2022
) ; This is not perfect.
946 (if (plist-get props
'escape-quoted
)
947 (error "escape-quoted is not supported: %S"
948 `(,name
,type
,doc-string
,props
)))
949 (let ((g0 (plist-get props
'charset-g0
))
950 (g1 (plist-get props
'charset-g1
))
951 (g2 (plist-get props
'charset-g2
))
952 (g3 (plist-get props
'charset-g3
))
955 (eq (cadr (assoc 'latin-jisx0201
956 (plist-get props
'input-charset-conversion
)))
958 (eq (cadr (assoc 'ascii
959 (plist-get props
'output-charset-conversion
)))
963 (eq (cadr (assoc 'japanese-jisx0208-1978
964 (plist-get props
'input-charset-conversion
)))
966 (eq (cadr (assoc 'japanese-jisx0208
967 (plist-get props
'output-charset-conversion
)))
968 'japanese-jisx0208-1978
))))
970 (if (plist-get props
'force-g0-on-output
)
974 (if (plist-get props
'force-g1-on-output
)
978 (if (plist-get props
'force-g2-on-output
)
982 (if (plist-get props
'force-g3-on-output
)
985 `(,name
2 ,mnemonic
,doc-string
987 ,(plist-get props
'short
)
988 ,(not (plist-get props
'no-ascii-eol
))
989 ,(not (plist-get props
'no-ascii-cntl
))
990 ,(plist-get props
'seven
)
992 ,(not (plist-get props
'lock-shift
))
995 ,(plist-get props
'no-iso6429
)
997 ,properties
,eol-type
)))
999 `(,name
3 ,mnemonic
,doc-string
() ,properties
,eol-type
))
1001 `(,name
4 ,mnemonic
,doc-string
1002 (,(plist-get props
'decode
) .
,(plist-get props
'encode
))
1003 ,properties
,eol-type
))
1005 (error "unsupported XEmacs style make-coding-style arguments: %S"
1006 `(,name
,type
,doc-string
,props
))))))
1008 (defun make-coding-system (coding-system type mnemonic doc-string
1013 "Define a new coding system CODING-SYSTEM (symbol).
1014 This function is provided for backward compatibility."
1015 (declare (obsolete define-coding-system
"23.1"))
1016 ;; For compatibility with XEmacs, we check the type of TYPE. If it
1017 ;; is a symbol, perhaps, this function is called with XEmacs-style
1018 ;; arguments. Here, try to transform that kind of arguments to
1021 (let ((args (transform-make-coding-system-args coding-system type
1022 mnemonic doc-string
)))
1023 (setq coding-system
(car args
)
1025 mnemonic
(nth 2 args
)
1026 doc-string
(nth 3 args
)
1028 properties
(nth 5 args
)
1029 eol-type
(nth 6 args
))))
1032 (cond ((eq type
0) 'emacs-mule
)
1033 ((eq type
1) 'shift-jis
)
1034 ((eq type
2) 'iso2022
)
1037 ((eq type
5) 'raw-text
)
1039 (error "Invalid coding system type: %s" type
))))
1042 (let ((plist nil
) key
)
1043 (dolist (elt properties
)
1044 (setq key
(car elt
))
1045 (cond ((eq key
'post-read-conversion
)
1046 (setq key
:post-read-conversion
))
1047 ((eq key
'pre-write-conversion
)
1048 (setq key
:pre-write-conversion
))
1049 ((eq key
'translation-table-for-decode
)
1050 (setq key
:decode-translation-table
))
1051 ((eq key
'translation-table-for-encode
)
1052 (setq key
:encode-translation-table
))
1053 ((eq key
'safe-charsets
)
1054 (setq key
:charset-list
))
1055 ((eq key
'mime-charset
)
1056 (setq key
:mime-charset
))
1057 ((eq key
'valid-codes
)
1058 (setq key
:valids
)))
1059 (setq plist
(plist-put plist key
(cdr elt
))))
1061 (setq properties
(plist-put properties
:mnemonic mnemonic
))
1062 (plist-put properties
:coding-type type
)
1063 (cond ((eq eol-type
0) (setq eol-type
'unix
))
1064 ((eq eol-type
1) (setq eol-type
'dos
))
1065 ((eq eol-type
2) (setq eol-type
'mac
))
1066 ((vectorp eol-type
) (setq eol-type nil
)))
1067 (plist-put properties
:eol-type eol-type
)
1071 (plist-put properties
:flags
1072 (list (and (or (consp (nth 0 flags
))
1073 (consp (nth 1 flags
))
1074 (consp (nth 2 flags
))
1075 (consp (nth 3 flags
))) 'designation
)
1076 (or (nth 4 flags
) 'long-form
)
1077 (and (nth 5 flags
) 'ascii-at-eol
)
1078 (and (nth 6 flags
) 'ascii-at-cntl
)
1079 (and (nth 7 flags
) '7-bit
)
1080 (and (nth 8 flags
) 'locking-shift
)
1081 (and (nth 9 flags
) 'single-shift
)
1082 (and (nth 10 flags
) 'use-roman
)
1083 (and (nth 11 flags
) 'use-oldjis
)
1084 (or (nth 12 flags
) 'direction
)
1085 (and (nth 13 flags
) 'init-at-bol
)
1086 (and (nth 14 flags
) 'designate-at-bol
)
1087 (and (nth 15 flags
) 'safe
)
1088 (and (nth 16 flags
) 'latin-extra
)))
1089 (plist-put properties
:designation
1090 (let ((vec (make-vector 4 nil
)))
1092 (let ((spec (nth i flags
)))
1094 (aset vec i
'(94 96))
1098 (setq spec
(append (delq t spec
) '(94 96))))
1099 (aset vec i spec
))))))
1103 (plist-put properties
:ccl-decoder
(car flags
))
1104 (plist-put properties
:ccl-encoder
(cdr flags
))))
1106 (apply 'define-coding-system coding-system doc-string properties
))
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
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 two 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) `set-buffer-file-coding-system' is called.
1140 ;; The cdr of this value is set to the specified coding system.
1141 ;; This variable is used for decoding in `revert-buffer' and encoding
1142 ;; in `select-safe-coding-system'.
1144 ;; When saving a buffer, if `buffer-file-coding-system-explicit' is
1145 ;; already non-nil, `basic-save-buffer-1' sets its CAR to the value of
1146 ;; `last-coding-system-used'. (It used to set it unconditionally, but
1147 ;; that seems unnecessary; see Bug#4533.)
1149 (defvar buffer-file-coding-system-explicit nil
1150 "The file coding system explicitly specified for the current buffer.
1151 The value is a cons of coding systems for reading (decoding) and
1153 Internal use only.")
1154 (make-variable-buffer-local 'buffer-file-coding-system-explicit
)
1155 (put 'buffer-file-coding-system-explicit
'permanent-local t
)
1157 (defun read-buffer-file-coding-system ()
1158 (let* ((bcss (find-coding-systems-region (point-min) (point-max)))
1160 (unless (equal bcss
'(undecided))
1161 (append '("dos" "unix" "mac")
1162 (delq nil
(mapcar (lambda (cs)
1163 (if (memq (coding-system-base cs
) bcss
)
1165 coding-system-list
)))))
1168 (completion-table-in-turn css-table coding-system-alist
)
1169 coding-system-alist
))
1171 (unless find-file-literally
1175 (goto-char (point-min))
1176 (funcall set-auto-coding-function
1177 (or buffer-file-name
"") (buffer-size))))))
1179 (let ((bfcs (default-value 'buffer-file-coding-system
)))
1180 (cons (and (or (equal bcss
'(undecided))
1181 (memq (coding-system-base bfcs
) bcss
))
1183 (mapcar (lambda (cs)
1184 (and (coding-system-p cs
)
1185 (coding-system-get cs
:mime-charset
)
1186 (or (equal bcss
'(undecided))
1187 (memq (coding-system-base cs
) bcss
))
1189 (coding-system-priority-list)))))
1191 (let ((current (coding-system-base buffer-file-coding-system
)))
1192 ;; Generally use as a default the first preferred coding-system
1193 ;; different from the current coding-system, except for
1194 ;; the case of auto-cs since choosing anything else is asking
1195 ;; for trouble (would lead to using a different coding
1196 ;; system than specified in the coding tag).
1199 (mapcar (lambda (cs)
1200 (if (eq current
(coding-system-base cs
))
1204 (completion-ignore-case t
)
1205 (completion-pcm--delim-wild-regex ; Let "u8" complete to "utf-8".
1206 (concat completion-pcm--delim-wild-regex
1207 "\\|\\([[:alpha:]]\\)[[:digit:]]"))
1208 (cs (completing-read
1209 (format "Coding system for saving file (default %s): " default
)
1211 nil t nil
'coding-system-history
1212 (if default
(symbol-name default
)))))
1213 (unless (zerop (length cs
)) (intern cs
))))
1215 (defun set-buffer-file-coding-system (coding-system &optional force nomodify
)
1216 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1217 This means that when you save the buffer, it will be converted
1218 according to CODING-SYSTEM. For a list of possible values of
1219 CODING-SYSTEM, use \\[list-coding-systems].
1221 If CODING-SYSTEM leaves the text conversion unspecified, or if it leaves
1222 the end-of-line conversion unspecified, FORCE controls what to do.
1223 If FORCE is nil, get the unspecified aspect (or aspects) from the buffer's
1224 previous `buffer-file-coding-system' value (if it is specified there).
1225 Otherwise, leave it unspecified.
1227 This marks the buffer modified so that the succeeding \\[save-buffer]
1228 surely saves the buffer with CODING-SYSTEM. From a program, if you
1229 don't want to mark the buffer modified, specify t for NOMODIFY.
1230 If you know exactly what coding system you want to use,
1231 just set the variable `buffer-file-coding-system' directly."
1233 (list (read-buffer-file-coding-system)
1234 current-prefix-arg
))
1235 (check-coding-system coding-system
)
1236 (if (and coding-system buffer-file-coding-system
(null force
))
1238 (merge-coding-systems coding-system buffer-file-coding-system
)))
1239 (when (called-interactively-p 'interactive
)
1240 ;; Check whether save would succeed, and jump to the offending char(s)
1242 (let ((css (find-coding-systems-region (point-min) (point-max))))
1243 (unless (or (eq (car css
) 'undecided
)
1244 (memq (coding-system-base coding-system
) css
))
1245 (setq coding-system
(select-safe-coding-system-interactively
1246 (point-min) (point-max) css
1247 (list coding-system
))))))
1248 (setq buffer-file-coding-system coding-system
)
1249 (if buffer-file-coding-system-explicit
1250 (setcdr buffer-file-coding-system-explicit coding-system
)
1251 (setq buffer-file-coding-system-explicit
(cons nil coding-system
)))
1253 (set-buffer-modified-p t
))
1254 (force-mode-line-update))
1256 (defun revert-buffer-with-coding-system (coding-system &optional force
)
1257 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1258 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1260 If CODING-SYSTEM leaves the text conversion unspecified, or if it leaves
1261 the end-of-line conversion unspecified, FORCE controls what to do.
1262 If FORCE is nil, get the unspecified aspect (or aspects) from the buffer's
1263 previous `buffer-file-coding-system' value (if it is specified there).
1264 Otherwise, determine it from the file contents as usual for visiting a file."
1265 (interactive "zCoding system for visited file (default nil): \nP")
1266 (check-coding-system coding-system
)
1267 (if (and coding-system buffer-file-coding-system
(null force
))
1269 (merge-coding-systems coding-system buffer-file-coding-system
)))
1270 (let ((coding-system-for-read coding-system
))
1273 (defun set-file-name-coding-system (coding-system)
1274 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1275 It actually just set the variable `file-name-coding-system' (which see)
1277 (interactive "zCoding system for file names (default nil): ")
1278 (check-coding-system coding-system
)
1279 (if (and coding-system
1280 (not (coding-system-get coding-system
:ascii-compatible-p
))
1281 (not (coding-system-get coding-system
:suitable-for-file-name
)))
1282 (error "%s is not suitable for file names" coding-system
))
1283 (setq file-name-coding-system coding-system
))
1285 (defvar default-terminal-coding-system nil
1286 "Default value for the terminal coding system.
1287 This is normally set according to the selected language environment.
1288 See also the command `set-terminal-coding-system'.")
1290 (defun set-terminal-coding-system (coding-system &optional terminal
)
1291 "Set coding system of terminal output to CODING-SYSTEM.
1292 All text output to TERMINAL will be encoded
1293 with the specified coding system.
1295 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1296 The default is determined by the selected language environment
1297 or by the previous use of this command.
1299 TERMINAL may be a terminal object, a frame, or nil for the
1300 selected frame's terminal. The setting has no effect on
1301 graphical terminals."
1303 (list (let ((default (if (and (not (terminal-coding-system))
1304 default-terminal-coding-system
)
1305 default-terminal-coding-system
)))
1307 (format "Coding system for terminal display (default %s): "
1310 (if (and (not coding-system
)
1311 (not (terminal-coding-system)))
1312 (setq coding-system default-terminal-coding-system
))
1314 (setq default-terminal-coding-system coding-system
))
1315 (set-terminal-coding-system-internal coding-system terminal
)
1316 (redraw-frame (selected-frame)))
1318 (defvar default-keyboard-coding-system nil
1319 "Default value of the keyboard coding system.
1320 This is normally set according to the selected language environment.
1321 See also the command `set-keyboard-coding-system'.")
1323 (defun set-keyboard-coding-system (coding-system &optional terminal
)
1324 "Set coding system for keyboard input on TERMINAL to CODING-SYSTEM.
1326 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1327 The default is determined by the selected language environment
1328 or by the previous use of this command.
1330 If CODING-SYSTEM is nil or the coding-type of CODING-SYSTEM is
1331 `raw-text', the decoding of keyboard input is disabled.
1333 TERMINAL may be a terminal object, a frame, or nil for the
1334 selected frame's terminal. The setting has no effect on
1335 graphical terminals."
1337 (list (let* ((coding (keyboard-coding-system nil
))
1338 (default (if (eq (coding-system-type coding
) 'raw-text
)
1339 default-keyboard-coding-system
)))
1341 (format "Coding system for keyboard input (default %s): "
1344 (let ((coding-type (coding-system-type coding-system
))
1346 (terminal-parameter terminal
'keyboard-coding-saved-meta-mode
)))
1347 (if (not (eq coding-type
'raw-text
))
1349 (if (not (or (coding-system-get coding-system
:suitable-for-keyboard
)
1350 (coding-system-get coding-system
:ascii-compatible-p
)))
1351 (error "Unsuitable coding system for keyboard: %s" coding-system
))
1352 (cond ((memq coding-type
'(charset utf-8 shift-jis big5 ccl
))
1353 (setq accept-8-bit t
))
1354 ((eq coding-type
'iso-2022
)
1355 (let ((flags (coding-system-get coding-system
:flags
)))
1356 (or (memq '7-bit flags
)
1357 (setq accept-8-bit t
))))
1359 (error "Unsupported coding system for keyboard: %s"
1364 (set-terminal-parameter terminal
1365 'keyboard-coding-saved-meta-mode
1366 (cons (nth 2 (current-input-mode))
1368 (set-input-meta-mode 8 terminal
))
1369 (when saved-meta-mode
1370 (set-input-meta-mode (car saved-meta-mode
) terminal
)
1371 (set-terminal-parameter terminal
1372 'keyboard-coding-saved-meta-mode
1374 ;; Avoid end-of-line conversion.
1376 (coding-system-change-eol-conversion coding-system
'unix
)))
1378 (when saved-meta-mode
1379 (set-input-meta-mode (car saved-meta-mode
) terminal
)
1380 (set-terminal-parameter terminal
1381 'keyboard-coding-saved-meta-mode
1383 (set-keyboard-coding-system-internal coding-system terminal
)
1384 (setq keyboard-coding-system coding-system
))
1386 (defcustom keyboard-coding-system nil
1387 "Specify coding system for keyboard input.
1388 If you set this on a terminal which can't distinguish Meta keys from
1389 8-bit characters, you will have to use ESC to type Meta characters.
1390 See Info node `Terminal Coding' and Info node `Unibyte Mode'.
1392 On non-windowing terminals, this is set from the locale by default.
1394 Setting this variable directly does not take effect;
1395 use either \\[customize] or \\[set-keyboard-coding-system]."
1396 :type
'(coding-system :tag
"Coding system")
1397 :link
'(info-link "(emacs)Terminal Coding")
1398 :link
'(info-link "(emacs)Unibyte Mode")
1399 :set
(lambda (symbol value
)
1400 ;; Don't load encoded-kb unnecessarily.
1401 (if (or value
(boundp 'encoded-kbd-setup-display
))
1402 (set-keyboard-coding-system value
)
1403 (set-default 'keyboard-coding-system nil
))) ; must initialize
1408 (defun set-buffer-process-coding-system (decoding encoding
)
1409 "Set coding systems for the process associated with the current buffer.
1410 DECODING is the coding system to be used to decode input from the process,
1411 ENCODING is the coding system to be used to encode output to the process.
1413 For a list of possible coding systems, use \\[list-coding-systems]."
1415 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1416 (let ((proc (get-buffer-process (current-buffer))))
1418 (error "No process")
1419 (check-coding-system decoding
)
1420 (check-coding-system encoding
)
1421 (set-process-coding-system proc decoding encoding
)))
1422 (force-mode-line-update))
1424 (defalias 'set-clipboard-coding-system
'set-selection-coding-system
)
1426 (defun set-selection-coding-system (coding-system)
1427 "Make CODING-SYSTEM used for communicating with other X clients.
1428 When sending or receiving text via cut_buffer, selection, and clipboard,
1429 the text is encoded or decoded by CODING-SYSTEM."
1430 (interactive "zCoding system for X selection: ")
1431 (check-coding-system coding-system
)
1432 (setq selection-coding-system coding-system
))
1434 ;; Coding system lastly specified by the command
1435 ;; set-next-selection-coding-system.
1436 (defvar last-next-selection-coding-system nil
)
1438 (defun set-next-selection-coding-system (coding-system)
1439 "Use CODING-SYSTEM for next communication with other window system clients.
1440 This setting is effective for the next communication only."
1442 (list (read-coding-system
1443 (if last-next-selection-coding-system
1444 (format "Coding system for the next selection (default %S): "
1445 last-next-selection-coding-system
)
1446 "Coding system for the next selection: ")
1447 last-next-selection-coding-system
)))
1449 (setq last-next-selection-coding-system coding-system
)
1450 (setq coding-system last-next-selection-coding-system
))
1451 (check-coding-system coding-system
)
1453 (setq next-selection-coding-system coding-system
))
1455 (defun set-coding-priority (arg)
1456 "Set priority of coding categories according to ARG.
1457 ARG is a list of coding categories ordered by priority.
1459 This function is provided for backward compatibility."
1460 (declare (obsolete set-coding-system-priority
"23.1"))
1461 (apply 'set-coding-system-priority
1462 (mapcar #'(lambda (x) (symbol-value x
)) arg
)))
1466 (defvar ctext-non-standard-encodings-alist
1468 '(("big5-0" big5
2 big5
)
1469 ("ISO8859-14" iso-8859-14
1 latin-iso8859-14
)
1470 ("ISO8859-15" iso-8859-15
1 latin-iso8859-15
)
1471 ("gbk-0" gbk
2 chinese-gbk
)
1472 ("koi8-r" koi8-r
1 koi8-r
)
1473 ("microsoft-cp1251" windows-1251
1 windows-1251
)))
1474 "Alist of non-standard encoding names vs the corresponding usages in CTEXT.
1476 It controls how extended segments of a compound text are handled
1477 by the coding system `compound-text-with-extensions'.
1479 Each element has the form (ENCODING-NAME CODING-SYSTEM N-OCTET CHARSET).
1481 ENCODING-NAME is an encoding name of an \"extended segment\".
1483 CODING-SYSTEM is the coding-system to encode (or decode) the
1484 characters into (or from) the extended segment.
1486 N-OCTET is the number of octets (bytes) that encodes a character
1487 in the segment. It can be 0 (meaning the number of octets per
1488 character is variable), 1, 2, 3, or 4.
1490 CHARSET is a character set containing characters that are encoded
1491 in the segment. It can be a list of character sets.
1493 On decoding CTEXT, all encoding names listed here are recognized.
1495 On encoding CTEXT, encoding names in the variable
1496 `ctext-non-standard-encodings' (which see) and in the information
1497 listed for the current language environment under the key
1498 `ctext-non-standard-encodings' are used.")
1500 (defvar ctext-non-standard-encodings nil
1501 "List of non-standard encoding names used in extended segments of CTEXT.
1502 Each element must be one of the names listed in the variable
1503 `ctext-non-standard-encodings-alist' (which see).")
1505 (defvar ctext-non-standard-encodings-regexp
1507 (string-to-multibyte
1509 ;; For non-standard encodings.
1510 "\\(\e%/[0-4][\200-\377][\200-\377]\\([^\002]+\\)\002\\)"
1512 ;; For UTF-8 encoding.
1513 "\\(\e%G[^\e]*\e%@\\)"))))
1515 ;; Functions to support "Non-Standard Character Set Encodings" defined
1516 ;; by the COMPOUND-TEXT spec. They also support "The UTF-8 encoding"
1517 ;; described in the section 7 of the documentation of COMPOUND-TEXT
1518 ;; distributed with XFree86.
1520 (defun ctext-post-read-conversion (len)
1521 "Decode LEN characters encoded as Compound Text with Extended Segments."
1522 ;; We don't need the following because it is expected that this
1523 ;; function is mainly used for decoding X selection which is not
1525 ;;(buffer-disable-undo) ; minimize consing due to insertions and deletions
1528 (narrow-to-region (point) (+ (point) len
))
1529 (let ((case-fold-search nil
)
1530 last-coding-system-used
1532 (decode-coding-region (point-min) (point-max) 'ctext
)
1533 (while (re-search-forward ctext-non-standard-encodings-regexp
1535 (setq pos
(match-beginning 0))
1536 (if (match-beginning 1)
1537 ;; ESC % / [0-4] M L --ENCODING-NAME-- \002 --BYTES--
1538 (let* ((M (multibyte-char-to-unibyte (char-after (+ pos
4))))
1539 (L (multibyte-char-to-unibyte (char-after (+ pos
5))))
1540 (encoding (match-string 2))
1541 (encoding-info (assoc-string
1543 ctext-non-standard-encodings-alist t
))
1544 (coding (if encoding-info
1545 (nth 1 encoding-info
)
1546 (setq encoding
(intern (downcase encoding
)))
1547 (and (coding-system-p encoding
)
1549 (setq bytes
(- (+ (* (- M
128) 128) (- L
128))
1550 (- (point) (+ pos
6))))
1552 (delete-region pos
(point))
1553 (forward-char bytes
)
1554 (decode-coding-region (- (point) bytes
) (point) coding
)))
1555 ;; ESC % G --UTF-8-BYTES-- ESC % @
1557 (delete-region pos
(+ pos
3))
1558 (decode-coding-region pos
(point) 'utf-8
))))
1559 (goto-char (point-min))
1560 (- (point-max) (point)))))
1562 (defvar ctext-standard-encodings
1563 '(ascii latin-jisx0201 katakana-jisx0201
1564 latin-iso8859-1 latin-iso8859-2 latin-iso8859-3 latin-iso8859-4
1565 greek-iso8859-7 arabic-iso8859-6 hebrew-iso8859-8 cyrillic-iso8859-5
1567 chinese-gb2312 japanese-jisx0208 korean-ksc5601
)
1568 "List of approved standard encodings (i.e. charsets) of X's Compound Text.
1569 Coding-system `compound-text-with-extensions' encodes a character
1570 belonging to any of those charsets using the normal ISO2022
1571 designation sequence unless the current language environment or
1572 the variable `ctext-non-standard-encodings' decide to use an extended
1573 segment of CTEXT for that character. See also the documentation
1574 of `ctext-non-standard-encodings-alist'.")
1576 ;; Return an alist of CHARSET vs CTEXT-USAGE-INFO generated from
1577 ;; `ctext-non-standard-encodings' and a list specified by the key
1578 ;; `ctext-non-standard-encodings' for the current language
1579 ;; environment. CTEXT-USAGE-INFO is one of the element of
1580 ;; `ctext-non-standard-encodings-alist' or nil. In the former case, a
1581 ;; character in CHARSET is encoded using extended segment. In the
1582 ;; latter case, a character in CHARSET is encoded using normal ISO2022
1583 ;; designation sequence. If a character is not in any of CHARSETs, it
1584 ;; is encoded using UTF-8 encoding extension.
1586 (defun ctext-non-standard-encodings-table ()
1587 (let* ((table (append ctext-non-standard-encodings
1589 (get-language-info current-language-environment
1590 'ctext-non-standard-encodings
))))
1594 (setq elt
(car tail
))
1595 (let* ((slot (assoc elt ctext-non-standard-encodings-alist
))
1596 (charset (nth 3 slot
)))
1597 (if (charsetp charset
)
1599 (cons (plist-get (charset-plist charset
) :base
) slot
))
1600 (setcar tail
(cons (car charset
) slot
))
1601 (dolist (cs (cdr charset
))
1603 (cons (cons (plist-get (charset-plist (car cs
)) :base
) slot
)
1605 (setq tail
(cdr tail
))))
1606 (setq tail
(cdr tail
))))
1609 (defun ctext-pre-write-conversion (from to
)
1610 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1612 If FROM is a string, generate a new temp buffer, insert the text,
1613 and convert it in the temporary buffer. Otherwise, convert
1616 ;; Setup a working buffer if necessary.
1617 (when (stringp from
)
1618 (set-buffer (generate-new-buffer " *temp"))
1619 (set-buffer-multibyte (multibyte-string-p from
))
1621 (setq from
(point-min) to
(point-max)))
1623 (narrow-to-region from to
)
1625 (let ((encoding-table (ctext-non-standard-encodings-table))
1626 (charset-list (sort-charsets
1627 (copy-sequence ctext-standard-encodings
)))
1628 (end-pos (make-marker))
1629 last-coding-system-used
1630 last-pos charset encoding-info
)
1631 (dolist (elt encoding-table
)
1632 (push (car elt
) charset-list
))
1633 (setq end-pos
(point-marker))
1634 (while (re-search-forward "[^\0-\177]+" nil t
)
1635 ;; Found a sequence of non-ASCII characters.
1636 (set-marker end-pos
(match-end 0))
1637 (goto-char (match-beginning 0))
1638 (setq last-pos
(point)
1639 charset
(char-charset (following-char) charset-list
))
1641 (while (and (< (point) end-pos
)
1642 (eq charset
(char-charset (following-char) charset-list
)))
1645 (if (setq encoding-info
(cdr (assq charset encoding-table
)))
1646 ;; Encode this range using an extended segment.
1647 (let ((encoding-name (car encoding-info
))
1648 (coding-system (nth 1 encoding-info
))
1649 (noctets (nth 2 encoding-info
))
1651 (encode-coding-region last-pos
(point) coding-system
)
1652 (setq len
(+ (length encoding-name
) 1
1653 (- (point) last-pos
)))
1654 ;; According to the spec of CTEXT, it is not
1655 ;; necessary to produce this extra designation
1656 ;; sequence, but some buggy application
1657 ;; (e.g. crxvt-gb) requires it.
1660 (goto-char last-pos
)
1661 (insert (format "\e%%/%d" noctets
))
1662 (insert-byte (+ (/ len
128) 128) 1)
1663 (insert-byte (+ (% len
128) 128) 1)
1664 (insert encoding-name
)
1666 ;; Encode this range as characters in CHARSET.
1667 (put-text-property last-pos
(point) 'charset charset
))
1668 ;; Encode this range using UTF-8 encoding extension.
1669 (encode-coding-region last-pos
(point) 'mule-utf-8
)
1671 (goto-char last-pos
)
1674 (goto-char (point-min)))))
1675 ;; Must return nil, as build_annotations_2 expects that.
1680 ;; TODO many elements of this list are also in inhibit-local-variables-regexps.
1681 (defcustom auto-coding-alist
1682 ;; .exe and .EXE are added to support archive-mode looking at DOS
1683 ;; self-extracting exe archives.
1684 (mapcar (lambda (arg) (cons (purecopy (car arg
)) (cdr arg
)))
1686 arc\\|zip\\|lzh\\|lha\\|zoo\\|[jew]ar\\|xpi\\|rar\\|7z\\|\
1687 ARC\\|ZIP\\|LZH\\|LHA\\|ZOO\\|[JEW]AR\\|XPI\\|RAR\\|7Z\\)\\'"
1688 . no-conversion-multibyte
)
1689 ("\\.\\(exe\\|EXE\\)\\'" . no-conversion
)
1690 ("\\.\\(sx[dmicw]\\|odt\\|tar\\|t[bg]z\\)\\'" . no-conversion
)
1691 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|xz\\|gpg\\)\\'" . no-conversion
)
1692 ("\\.\\(jpe?g\\|png\\|gif\\|tiff?\\|p[bpgn]m\\)\\'" . no-conversion
)
1693 ("\\.pdf\\'" . no-conversion
)
1694 ("/#[^/]+#\\'" . utf-8-emacs-unix
)))
1695 "Alist of filename patterns vs corresponding coding systems.
1696 Each element looks like (REGEXP . CODING-SYSTEM).
1697 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1699 The settings in this alist take priority over `coding:' tags
1700 in the file (see the function `set-auto-coding')
1701 and the contents of `file-coding-system-alist'."
1702 :version
"24.1" ; added xz
1705 :type
'(repeat (cons (regexp :tag
"File name regexp")
1706 (symbol :tag
"Coding system"))))
1708 (defcustom auto-coding-regexp-alist
1709 (mapcar (lambda (arg) (cons (purecopy (car arg
)) (cdr arg
)))
1710 '(("\\`BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion
)
1711 ("\\`\xFE\xFF" . utf-16be-with-signature
)
1712 ("\\`\xFF\xFE" . utf-16le-with-signature
)
1713 ("\\`\xEF\xBB\xBF" . utf-8-with-signature
)
1714 ("\\`;ELC\024\0\0\0" . emacs-mule
))) ; Emacs 20-compiled
1715 "Alist of patterns vs corresponding coding systems.
1716 Each element looks like (REGEXP . CODING-SYSTEM).
1717 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1719 The settings in this alist take priority over `coding:' tags
1720 in the file (see the function `set-auto-coding')
1721 and the contents of `file-coding-system-alist'."
1724 :type
'(repeat (cons (regexp :tag
"Regexp")
1725 (symbol :tag
"Coding system"))))
1727 (defun auto-coding-regexp-alist-lookup (from to
)
1728 "Lookup `auto-coding-regexp-alist' for the contents of the current buffer.
1729 The value is a coding system is specified for the region FROM and TO,
1733 (let ((alist auto-coding-regexp-alist
)
1735 (while (and alist
(not coding-system
))
1736 (let ((regexp (car (car alist
))))
1737 (if enable-multibyte-characters
1738 (setq regexp
(string-to-multibyte regexp
)))
1739 (if (re-search-forward regexp to t
)
1740 (setq coding-system
(cdr (car alist
)))
1741 (setq alist
(cdr alist
)))))
1744 ;; See the bottom of this file for built-in auto coding functions.
1745 (defcustom auto-coding-functions
'(sgml-xml-auto-coding-function
1746 sgml-html-meta-auto-coding-function
)
1747 "A list of functions which attempt to determine a coding system.
1749 Each function in this list should be written to operate on the
1750 current buffer, but should not modify it in any way. The buffer
1751 will contain undecoded text of parts of the file. Each function
1752 should take one argument, SIZE, which says how many characters
1753 \(starting from point) it should look at.
1755 If one of these functions succeeds in determining a coding
1756 system, it should return that coding system. Otherwise, it
1759 If a file has a `coding:' tag, that takes precedence over these
1760 functions, so they won't be called at all."
1763 :type
'(repeat function
))
1765 (defvar set-auto-coding-for-load nil
1766 "Non-nil means respect a \"unibyte: t\" entry in file local variables.
1767 Emacs binds this variable to t when loading or byte-compiling Emacs Lisp
1770 (defun auto-coding-alist-lookup (filename)
1771 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1772 (let ((alist auto-coding-alist
)
1773 (case-fold-search (memq system-type
'(windows-nt ms-dos cygwin
)))
1775 (while (and alist
(not coding-system
))
1776 (if (string-match (car (car alist
)) filename
)
1777 (setq coding-system
(cdr (car alist
)))
1778 (setq alist
(cdr alist
))))
1781 (put 'enable-character-translation
'permanent-local t
)
1782 (put 'enable-character-translation
'safe-local-variable
'booleanp
)
1784 (defun find-auto-coding (filename size
)
1785 "Find a coding system for a file FILENAME of which SIZE bytes follow point.
1786 These bytes should include at least the first 1k of the file
1787 and the last 3k of the file, but the middle may be omitted.
1789 The function checks FILENAME against the variable `auto-coding-alist'.
1790 If FILENAME doesn't match any entries in the variable, it checks the
1791 contents of the current buffer following point against
1792 `auto-coding-regexp-alist'. If no match is found, it checks for a
1793 `coding:' tag in the first one or two lines following point. If no
1794 `coding:' tag is found, it checks any local variables list in the last
1795 3K bytes out of the SIZE bytes. Finally, if none of these methods
1796 succeed, it checks to see if any function in `auto-coding-functions'
1799 If a coding system is specified, the return value is a cons
1800 \(CODING . SOURCE), where CODING is the specified coding system and
1801 SOURCE is a symbol `auto-coding-alist', `auto-coding-regexp-alist',
1802 `:coding', or `auto-coding-functions' indicating by what CODING is
1803 specified. Note that the validity of CODING is not checked;
1804 it's the caller's responsibility to check it.
1806 If nothing is specified, the return value is nil."
1807 (or (let ((coding-system (auto-coding-alist-lookup filename
)))
1809 (cons coding-system
'auto-coding-alist
)))
1810 ;; Try using `auto-coding-regexp-alist'.
1811 (let ((coding-system (auto-coding-regexp-alist-lookup (point)
1814 (cons coding-system
'auto-coding-regexp-alist
)))
1815 (let* ((case-fold-search t
)
1816 (head-start (point))
1817 (head-end (+ head-start
(min size
1024)))
1818 (tail-start (+ head-start
(max (- size
3072) 0)))
1819 (tail-end (+ head-start size
))
1820 coding-system head-found tail-found pos char-trans
)
1821 ;; Try a short cut by searching for the string "coding:"
1822 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1823 (setq head-found
(or (search-forward "coding:" head-end t
)
1824 (search-forward "unibyte:" head-end t
)
1825 (search-forward "enable-character-translation:"
1827 (if (and head-found
(> head-found tail-start
))
1828 ;; Head and tail are overlapped.
1829 (setq tail-found head-found
)
1830 (goto-char tail-start
)
1831 (setq tail-found
(or (search-forward "coding:" tail-end t
)
1832 (search-forward "unibyte:" tail-end t
)
1833 (search-forward "enable-character-translation:"
1836 ;; At first check the head.
1838 (goto-char head-start
)
1839 (setq head-end
(set-auto-mode-1))
1840 (setq head-start
(point))
1841 (when (and head-end
(< head-found head-end
))
1842 (goto-char head-start
)
1843 (when (and set-auto-coding-for-load
1845 "\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1847 (display-warning 'mule
1848 (format "\"unibyte: t\" (in %s) is obsolete; \
1849 use \"coding: 'raw-text\" instead."
1850 (file-relative-name filename
))
1852 (setq coding-system
'raw-text
))
1853 (when (and (not coding-system
)
1855 "\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1857 (setq coding-system
(intern (match-string 2))))
1858 (when (re-search-forward
1859 "\\(.*;\\)?[ \t]*enable-character-translation:[ \t]*\\([^ ;]+\\)"
1861 (setq char-trans
(match-string 2)))))
1863 ;; If no coding: tag in the head, check the tail.
1864 ;; Here we must pay attention to the case that the end-of-line
1865 ;; is just "\r" and we can't use "^" nor "$" in regexp.
1866 (when (and tail-found
(or (not coding-system
) (not char-trans
)))
1867 (goto-char tail-start
)
1868 (re-search-forward "[\r\n]\^L" tail-end t
)
1869 (if (re-search-forward
1870 "[\r\n]\\([^[\r\n]*\\)[ \t]*Local Variables:[ \t]*\\([^\r\n]*\\)[\r\n]"
1872 ;; The prefix is what comes before "local variables:" in its
1873 ;; line. The suffix is what comes after "local variables:"
1875 (let* ((prefix (regexp-quote (match-string 1)))
1876 (suffix (regexp-quote (match-string 2)))
1880 ;; N.B. without the \n below, the regexp can
1882 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1887 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1892 "[ \t]*enable-character-translation[ \t]*:[ \t]*\\([^ \t\r\n]+\\)[ \t]*"
1895 (concat "[\r\n]" prefix
"[ \t]*End *:[ \t]*" suffix
1898 (forward-char -
1) ; skip back \r or \n.
1899 (re-search-forward re-end tail-end
'move
)
1900 (setq tail-end
(point))
1902 (when (and set-auto-coding-for-load
1903 (re-search-forward re-unibyte tail-end t
))
1904 (display-warning 'mule
"`unibyte: t' is obsolete; \
1905 use \"coding: 'raw-text\" instead." :warning
)
1906 (setq coding-system
'raw-text
))
1907 (when (and (not coding-system
)
1908 (re-search-forward re-coding tail-end t
))
1909 (setq coding-system
(intern (match-string 1))))
1910 (when (and (not char-trans
)
1911 (re-search-forward re-char-trans tail-end t
))
1912 (setq char-trans
(match-string 1))))))
1914 ;; If the coding-system name ends with "!", remove it and
1915 ;; set char-trans to "nil".
1916 (let ((name (symbol-name coding-system
)))
1917 (if (= (aref name
(1- (length name
))) ?
!)
1918 (setq coding-system
(intern (substring name
0 -
1))
1919 char-trans
"nil"))))
1920 (when (and char-trans
1921 (not (setq char-trans
(intern char-trans
))))
1922 (make-local-variable 'enable-character-translation
)
1923 (setq enable-character-translation nil
))
1925 (cons coding-system
:coding
)))
1926 ;; Finally, try all the `auto-coding-functions'.
1927 (let ((funcs auto-coding-functions
)
1928 (coding-system nil
))
1929 (while (and funcs
(not coding-system
))
1930 (setq coding-system
(condition-case e
1932 (goto-char (point-min))
1933 (funcall (pop funcs
) size
))
1936 (cons coding-system
'auto-coding-functions
)))))
1938 (defun set-auto-coding (filename size
)
1939 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1940 See `find-auto-coding' for how the coding system is found.
1941 Return nil if an invalid coding system is found.
1943 The variable `set-auto-coding-function' (which see) is set to this
1944 function by default."
1945 (let ((found (find-auto-coding filename size
)))
1946 (if (and found
(coding-system-p (car found
)))
1949 (setq set-auto-coding-function
'set-auto-coding
)
1951 (defun after-insert-file-set-coding (inserted &optional visit
)
1952 "Set `buffer-file-coding-system' of current buffer after text is inserted.
1953 INSERTED is the number of characters that were inserted, as figured
1954 in the situation before this function. Return the number of characters
1955 inserted, as figured in the situation after. The two numbers can be
1956 different if the buffer has become unibyte.
1957 The optional second arg VISIT non-nil means that we are visiting a file."
1959 coding-system-for-read
1960 (not (eq coding-system-for-read
'auto-save-coding
)))
1961 (setq buffer-file-coding-system-explicit
1962 (cons coding-system-for-read nil
)))
1963 (if last-coding-system-used
1964 (let ((coding-system
1965 (find-new-buffer-file-coding-system last-coding-system-used
)))
1967 (setq buffer-file-coding-system coding-system
))))
1970 ;; The coding-spec and eol-type of coding-system returned is decided
1971 ;; independently in the following order.
1972 ;; 1. That of buffer-file-coding-system locally bound.
1973 ;; 2. That of CODING.
1975 (defun find-new-buffer-file-coding-system (coding)
1976 "Return a coding system for a buffer when a file of CODING is inserted.
1977 The local variable `buffer-file-coding-system' of the current buffer
1978 is set to the returned value.
1979 Return nil if there's no need to set `buffer-file-coding-system'."
1980 (let (local-coding local-eol
1981 found-coding found-eol
1984 ;; Nothing found about coding.
1987 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1988 ;; and LOCAL-CODING.
1989 (setq local-eol
(coding-system-eol-type buffer-file-coding-system
))
1990 (if (null (numberp local-eol
))
1991 ;; But eol-type is not yet set.
1992 (setq local-eol nil
))
1993 (if (and buffer-file-coding-system
1994 (not (eq (coding-system-type buffer-file-coding-system
)
1996 (setq local-coding
(coding-system-base buffer-file-coding-system
)))
1998 (if (and (local-variable-p 'buffer-file-coding-system
)
1999 local-eol local-coding
)
2000 ;; The current buffer has already set full coding-system, we
2001 ;; had better not change it.
2004 (setq found-eol
(coding-system-eol-type coding
))
2005 (if (null (numberp found-eol
))
2006 ;; But eol-type is not found.
2007 ;; If EOL conversions are inhibited, force unix eol-type.
2008 (setq found-eol
(if inhibit-eol-conversion
0)))
2009 (setq found-coding
(coding-system-base coding
))
2011 (if (and (not found-eol
) (eq found-coding
'undecided
))
2012 ;; No valid coding information found.
2015 ;; Some coding information (eol or text) found.
2017 ;; The local setting takes precedence over the found one.
2018 (setq new-coding
(if (local-variable-p 'buffer-file-coding-system
)
2019 (or local-coding found-coding
)
2020 (or found-coding local-coding
)))
2021 (setq new-eol
(if (local-variable-p 'buffer-file-coding-system
)
2022 (or local-eol found-eol
)
2023 (or found-eol local-eol
)))
2025 (let ((eol-type (coding-system-eol-type new-coding
)))
2026 (if (and (numberp new-eol
) (vectorp eol-type
))
2027 (aref eol-type new-eol
)
2030 (defun modify-coding-system-alist (target-type regexp coding-system
)
2031 "Modify one of look up tables for finding a coding system on I/O operation.
2032 There are three of such tables, `file-coding-system-alist',
2033 `process-coding-system-alist', and `network-coding-system-alist'.
2035 TARGET-TYPE specifies which of them to modify.
2036 If it is `file', it affects `file-coding-system-alist' (which see).
2037 If it is `process', it affects `process-coding-system-alist' (which see).
2038 If it is `network', it affects `network-coding-system-alist' (which see).
2040 REGEXP is a regular expression matching a target of I/O operation.
2041 The target is a file name if TARGET-TYPE is `file', a program name if
2042 TARGET-TYPE is `process', or a network service name or a port number
2043 to connect to if TARGET-TYPE is `network'.
2045 CODING-SYSTEM is a coding system to perform code conversion on the I/O
2046 operation, or a cons cell (DECODING . ENCODING) specifying the coding
2047 systems for decoding and encoding respectively, or a function symbol
2048 which, when called, returns such a cons cell."
2049 (or (memq target-type
'(file process network
))
2050 (error "Invalid target type: %s" target-type
))
2051 (or (stringp regexp
)
2052 (and (eq target-type
'network
) (integerp regexp
))
2053 (error "Invalid regular expression: %s" regexp
))
2054 (if (symbolp coding-system
)
2055 (if (not (fboundp coding-system
))
2057 (check-coding-system coding-system
)
2058 (setq coding-system
(cons coding-system coding-system
))))
2059 (check-coding-system (car coding-system
))
2060 (check-coding-system (cdr coding-system
)))
2061 (cond ((eq target-type
'file
)
2062 (let ((slot (assoc regexp file-coding-system-alist
)))
2064 (setcdr slot coding-system
)
2065 (setq file-coding-system-alist
2066 (cons (cons regexp coding-system
)
2067 file-coding-system-alist
)))))
2068 ((eq target-type
'process
)
2069 (let ((slot (assoc regexp process-coding-system-alist
)))
2071 (setcdr slot coding-system
)
2072 (setq process-coding-system-alist
2073 (cons (cons regexp coding-system
)
2074 process-coding-system-alist
)))))
2076 (let ((slot (assoc regexp network-coding-system-alist
)))
2078 (setcdr slot coding-system
)
2079 (setq network-coding-system-alist
2080 (cons (cons regexp coding-system
)
2081 network-coding-system-alist
)))))))
2083 (defun decode-coding-inserted-region (from to filename
2084 &optional visit beg end replace
)
2085 "Decode the region between FROM and TO as if it is read from file FILENAME.
2086 The idea is that the text between FROM and TO was just inserted somehow.
2087 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
2088 of the function `insert-file-contents'.
2089 Part of the job of this function is setting `buffer-undo-list' appropriately."
2092 (let ((coding coding-system-for-read
)
2095 ;; Temporarily turn off undo recording, if we're decoding the
2096 ;; text of a visited file.
2097 (setq buffer-undo-list t
)
2098 ;; Otherwise, if we can recognize the undo elt for the insertion,
2099 ;; remove it and get ready to replace it later.
2100 ;; In the mean time, turn off undo recording.
2101 (let ((last (car-safe buffer-undo-list
)))
2102 (if (and (consp last
) (eql (car last
) from
) (eql (cdr last
) to
))
2103 (setq undo-list-saved
(cdr buffer-undo-list
)
2104 buffer-undo-list t
))))
2105 (narrow-to-region from to
)
2106 (goto-char (point-min))
2108 (setq coding
(funcall set-auto-coding-function
2109 filename
(- (point-max) (point-min)))))
2111 (setq coding
(car (find-operation-coding-system
2112 'insert-file-contents
2113 (cons filename
(current-buffer))
2114 visit beg end replace
))))
2115 (if (coding-system-p coding
)
2116 (or enable-multibyte-characters
2118 (coding-system-change-text-conversion coding
'raw-text
)))
2121 (decode-coding-region (point-min) (point-max) coding
)
2122 (setq last-coding-system-used coding
))
2123 ;; If we're decoding the text of a visited file,
2124 ;; the undo list should start out empty.
2126 (setq buffer-undo-list nil
)
2127 ;; If we decided to replace the undo entry for the insertion,
2130 (setq buffer-undo-list
2131 (cons (cons from
(point-max)) undo-list-saved
))))))))
2133 (defun recode-region (start end new-coding coding
)
2134 "Re-decode the region (previously decoded by CODING) by NEW-CODING."
2136 (list (region-beginning) (region-end)
2137 (read-coding-system "Text was really in: ")
2138 (let ((coding (or buffer-file-coding-system last-coding-system-used
)))
2140 (concat "But was interpreted as"
2141 (if coding
(format " (default %S): " coding
) ": "))
2143 (or (and new-coding coding
)
2144 (error "Coding system not specified"))
2145 ;; Check it before we encode the region.
2146 (check-coding-system new-coding
)
2148 (narrow-to-region start end
)
2149 (encode-coding-region (point-min) (point-max) coding
)
2150 (decode-coding-region (point-min) (point-max) new-coding
))
2151 (if (region-active-p)
2154 (defun make-translation-table (&rest args
)
2155 "Make a translation table from arguments.
2156 A translation table is a char table intended for character
2157 translation in CCL programs.
2159 Each argument is a list of elements of the form (FROM . TO), where FROM
2160 is a character to be translated to TO.
2162 The arguments and forms in each argument are processed in the given
2163 order, and if a previous form already translates TO to some other
2164 character, say TO-ALT, FROM is also translated to TO-ALT."
2165 (let ((table (make-char-table 'translation-table
))
2169 (let ((from (car elt
))
2171 to-alt rev-from rev-to
)
2172 ;; If we have already translated TO to TO-ALT, FROM should
2173 ;; also be translated to TO-ALT.
2174 (if (setq to-alt
(aref table to
))
2176 (aset table from to
)
2177 ;; If we have already translated some chars to FROM, they
2178 ;; should also be translated to TO.
2179 (when (setq rev-from
(assq from revlist
))
2180 (dolist (elt (cdr rev-from
))
2181 (aset table elt to
))
2182 (setq revlist
(delq rev-from revlist
)
2183 rev-from
(cdr rev-from
)))
2184 ;; Now update REVLIST.
2185 (setq rev-to
(assq to revlist
))
2187 (setcdr rev-to
(cons from
(cdr rev-to
)))
2188 (setq rev-to
(list to from
)
2189 revlist
(cons rev-to revlist
)))
2191 (setcdr rev-to
(append rev-from
(cdr rev-to
)))))))
2192 ;; Return TABLE just created.
2193 (set-char-table-extra-slot table
1 1)
2196 (defun make-translation-table-from-vector (vec)
2197 "Make translation table from decoding vector VEC.
2198 VEC is an array of 256 elements to map unibyte codes to multibyte
2199 characters. Elements may be nil for undefined code points."
2200 (let ((table (make-char-table 'translation-table
))
2201 (rev-table (make-char-table 'translation-table
))
2204 (setq ch
(aref vec i
))
2208 (aset rev-table ch i
))))
2209 (set-char-table-extra-slot table
0 rev-table
)
2210 (set-char-table-extra-slot table
1 1)
2211 (set-char-table-extra-slot rev-table
1 1)
2214 (defun make-translation-table-from-alist (alist)
2215 "Make translation table from N<->M mapping in ALIST.
2216 ALIST is an alist, each element has the form (FROM . TO).
2217 FROM and TO are a character or a vector of characters.
2218 If FROM is a character, that character is translated to TO.
2219 If FROM is a vector of characters, that sequence is translated to TO.
2220 The first extra-slot of the value is a translation table for reverse mapping."
2221 (let ((tables (vector (make-char-table 'translation-table
)
2222 (make-char-table 'translation-table
)))
2223 table max-lookup from to idx val
)
2225 (setq table
(aref tables i
))
2229 (setq from
(car elt
) to
(cdr elt
))
2230 (setq from
(cdr elt
) to
(car elt
)))
2231 (if (characterp from
)
2233 (setq idx
(aref from
0)
2234 max-lookup
(max max-lookup
(length from
))))
2235 (setq val
(aref table idx
))
2239 (setq val
(list (cons (vector idx
) val
))))
2240 (if (characterp from
)
2241 (setq from
(vector from
)))
2242 (setq val
(nconc val
(list (cons from to
)))))
2243 (if (characterp from
)
2245 (setq val
(list (cons from to
)))))
2246 (aset table idx val
))
2247 (set-char-table-extra-slot table
1 max-lookup
))
2248 (set-char-table-extra-slot (aref tables
0) 0 (aref tables
1))
2251 (defun define-translation-table (symbol &rest args
)
2252 "Define SYMBOL as the name of translation table made by ARGS.
2253 This sets up information so that the table can be used for
2254 translations in a CCL program.
2256 If the first element of ARGS is a char-table whose purpose is
2257 `translation-table', just define SYMBOL to name it. (Note that this
2258 function does not bind SYMBOL.)
2260 Any other ARGS should be suitable as arguments of the function
2261 `make-translation-table' (which see).
2263 This function sets properties `translation-table' and
2264 `translation-table-id' of SYMBOL to the created table itself and the
2265 identification number of the table respectively. It also registers
2266 the table in `translation-table-vector'."
2267 (let ((table (if (and (char-table-p (car args
))
2268 (eq (char-table-subtype (car args
))
2269 'translation-table
))
2271 (apply 'make-translation-table args
)))
2272 (len (length translation-table-vector
))
2275 (put symbol
'translation-table table
)
2278 (setq translation-table-vector
2279 (vconcat translation-table-vector
(make-vector len nil
))))
2280 (let ((slot (aref translation-table-vector id
)))
2282 (eq (car slot
) symbol
))
2284 (aset translation-table-vector id
(cons symbol table
))
2286 (setq id
(1+ id
)))))
2287 (put symbol
'translation-table-id id
)
2290 (defun translate-region (start end table
)
2291 "From START to END, translate characters according to TABLE.
2292 TABLE is a string or a char-table.
2293 If TABLE is a string, the Nth character in it is the mapping
2294 for the character with code N.
2295 If TABLE is a char-table, the element for character N is the mapping
2296 for the character with code N.
2297 It returns the number of characters changed."
2299 (list (region-beginning)
2302 (dotimes (i (length translation-table-vector
))
2303 (if (consp (aref translation-table-vector i
))
2304 (push (list (symbol-name
2305 (car (aref translation-table-vector i
)))) l
)))
2307 (error "No translation table defined"))
2309 (setq table
(completing-read "Translation table: " l nil t
)))
2312 (let ((val (get table
'translation-table
)))
2313 (or (char-table-p val
)
2314 (error "Invalid translation table name: %s" table
))
2316 (translate-region-internal start end table
))
2318 (defmacro with-category-table
(table &rest body
)
2319 "Execute BODY like `progn' with TABLE the current category table.
2320 The category table of the current buffer is saved, BODY is evaluated,
2321 then the saved table is restored, even in case of an abnormal exit.
2322 Value is what BODY returns."
2323 (declare (indent 1) (debug t
))
2324 (let ((old-table (make-symbol "old-table"))
2325 (old-buffer (make-symbol "old-buffer")))
2326 `(let ((,old-table
(category-table))
2327 (,old-buffer
(current-buffer)))
2330 (set-category-table ,table
)
2332 (with-current-buffer ,old-buffer
2333 (set-category-table ,old-table
))))))
2335 (defun define-translation-hash-table (symbol table
)
2336 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2338 Analogous to `define-translation-table', but updates
2339 `translation-hash-table-vector' and the table is for use in the CCL
2340 `lookup-integer' and `lookup-character' functions."
2341 (unless (and (symbolp symbol
)
2342 (hash-table-p table
))
2343 (error "Bad args to define-translation-hash-table"))
2344 (let ((len (length translation-hash-table-vector
))
2347 (put symbol
'translation-hash-table table
)
2350 (setq translation-hash-table-vector
2351 (vconcat translation-hash-table-vector
[nil])))
2352 (let ((slot (aref translation-hash-table-vector id)))
2354 (eq (car slot) symbol))
2356 (aset translation-hash-table-vector id (cons symbol table))
2358 (setq id (1+ id)))))
2359 (put symbol 'translation-hash-table-id id)
2362 ;;; Initialize some variables.
2364 (put 'use-default-ascent 'char-table-extra-slots 0)
2365 (setq use-default-ascent (make-char-table 'use-default-ascent))
2366 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2367 (setq ignore-relative-composition
2368 (make-char-table 'ignore-relative-composition))
2370 ;;; Built-in auto-coding-functions:
2372 (defun sgml-xml-auto-coding-function (size)
2373 "Determine whether the buffer is XML, and if so, its encoding.
2374 This function is intended to be added to `auto-coding-functions'."
2375 (setq size (+ (point) size))
2376 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" size t)
2377 (let ((end (save-excursion
2379 (re-search-forward "[\"']\\s-*\\?>" size t))))
2381 (if (re-search-forward "encoding=[\"']\\(.+?\\)[\"']" end t)
2382 (let* ((match (match-string 1))
2383 (sym (intern (downcase match))))
2384 (if (coding-system-p sym)
2386 (message "Warning: unknown coding system \"%s\"" match)
2388 ;; Files without an encoding tag should be UTF-8. But users
2389 ;; may be naive about encodings, and have saved the file from
2390 ;; another editor that does not help them get the encoding right.
2391 ;; Detect the encoding and warn the user if it is detected as
2392 ;; something other than UTF-8.
2394 (with-coding-priority '(utf-8)
2396 (detect-coding-region (point-min) size t)))))
2397 ;; Pure ASCII always comes back as undecided.
2398 (if (memq detected '(utf-8 undecided))
2400 (warn "File contents detected as %s.
2401 Consider adding an encoding attribute to the xml declaration,
2402 or saving as utf-8, as mandated by the xml specification." detected)
2405 (defun sgml-html-meta-auto-coding-function (size)
2406 "If the buffer has an HTML meta tag, use it to determine encoding.
2407 This function is intended to be added to `auto-coding-functions'."
2408 (let ((case-fold-search t))
2409 (setq size (min (+ (point) size)
2411 ;; Limit the search by the end of the HTML header.
2412 (or (search-forward "</head>" (+ (point) size) t)
2413 ;; In case of no header, search only 10 lines.
2416 ;; Make sure that the buffer really contains an HTML document, by
2417 ;; checking that it starts with a doctype or a <HTML> start tag
2418 ;; (allowing for whitespace at bob). Note: 'DOCTYPE NETSCAPE' is
2419 ;; useful for Mozilla bookmark files.
2420 (when (and (re-search-forward "\\`[[:space:]\n]*\\(<!doctype[[:space:]\n]+\\(html\\|netscape\\)\\|<html\\)" size t)
2421 (re-search-forward "<meta\\s-+\\(http-equiv=[\"']?content-type[\"']?\\s-+content=[\"']text/\\sw+;\\s-*\\)?charset=[\"']?\\(.+?\\)[\"'\\s-/>]" size t))
2422 (let* ((match (match-string 2))
2423 (sym (intern (downcase match))))
2424 (if (coding-system-p sym)
2426 (message "Warning: unknown coding system \"%s\"" match)
2429 (defun xml-find-file-coding-system (args)
2430 "Determine the coding system of an XML file without a declaration.
2431 Strictly speaking, the file should be utf-8, but mistakes are
2432 made, and there are genuine cases where XML fragments are saved,
2433 with the encoding properly specified in a master document, or
2434 added by processing software."
2435 (if (eq (car args) 'insert-file-contents)
2437 (with-coding-priority '(utf-8)
2439 (detect-coding-region (point-min) (point-max) t)))))
2440 ;; Pure ASCII always comes back as undecided.
2442 ((memq detected '(utf-8 undecided))
2444 ((eq detected 'utf-16le-with-signature) 'utf-16le-with-signature)
2445 ((eq detected 'utf-16be-with-signature) 'utf-16be-with-signature)
2447 (warn "File contents detected as %s.
2448 Consider adding an xml declaration with the encoding specified,
2449 or saving as utf-8, as mandated by the xml specification." detected)
2451 ;; Don't interfere with the user's wishes for saving the buffer.
2452 ;; We did what we could when the buffer was created to ensure the
2453 ;; correct encoding was used, or the user was warned, so any
2454 ;; non-conformity here is deliberate on the part of the user.
2460 ;;; mule.el ends here