1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
5 ;; Copyright (C) 2001, 2002 Free Software Foundation, Inc.
7 ;; Keywords: mule, multilingual, character set, coding system
9 ;; This file is part of GNU Emacs.
11 ;; GNU Emacs is free software; you can redistribute it and/or modify
12 ;; it under the terms of the GNU General Public License as published by
13 ;; the Free Software Foundation; either version 2, or (at your option)
16 ;; GNU Emacs is distributed in the hope that it will be useful,
17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 ;; GNU General Public License for more details.
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with GNU Emacs; see the file COPYING. If not, write to the
23 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
24 ;; Boston, MA 02111-1307, USA.
30 (defconst mule-version
"5.0 (SAKAKI)" "\
31 Version number and name of this version of MULE (multilingual environment).")
33 (defconst mule-version-date
"1999.12.7" "\
34 Distribution date of this version of MULE (multilingual environment).")
36 (defun load-with-code-conversion (fullname file
&optional noerror nomessage
)
37 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
38 The file contents are decoded before evaluation if necessary.
39 If optional second arg NOERROR is non-nil,
40 report no error if FILE doesn't exist.
41 Print messages at start and end of loading unless
42 optional third arg NOMESSAGE is non-nil.
43 Return t if file exists."
44 (if (null (file-readable-p fullname
))
46 (signal 'file-error
(list "Cannot open load file" file
)))
47 ;; Read file with code conversion, and then eval.
49 ;; To avoid any autoloading, set default-major-mode to
51 ;; So that we don't get completely screwed if the
52 ;; file is encoded in some complicated character set,
53 ;; read it with real decoding, as a multibyte buffer,
54 ;; even if this is a --unibyte Emacs session.
55 (let ((default-major-mode 'fundamental-mode
)
56 (default-enable-multibyte-characters t
))
57 ;; We can't use `generate-new-buffer' because files.el
59 (get-buffer-create (generate-new-buffer-name " *load*"))))
61 (source (save-match-data (string-match "\\.el\\'" fullname
))))
64 (message "Loading %s (source)..." file
)
65 (message "Loading %s..." file
)))
67 (push file preloaded-file-list
))
69 (let ((load-file-name fullname
)
70 (set-auto-coding-for-load t
)
71 (inhibit-file-name-operation nil
))
74 (insert-file-contents fullname
)
75 ;; If the loaded file was inserted with no-conversion or
76 ;; raw-text coding system, make the buffer unibyte.
77 ;; Otherwise, eval-buffer might try to interpret random
78 ;; binary junk as multibyte characters.
79 (if (and enable-multibyte-characters
80 (or (eq (coding-system-type last-coding-system-used
) 5)
81 (eq last-coding-system-used
'no-conversion
)))
82 (set-buffer-multibyte nil
))
83 ;; Make `kill-buffer' quiet.
84 (set-buffer-modified-p nil
))
85 ;; Have the original buffer current while we eval.
86 (eval-buffer buffer nil file
87 ;; If this Emacs is running with --unibyte,
88 ;; convert multibyte strings to unibyte
89 ;; after reading them.
90 ;; (not default-enable-multibyte-characters)
93 (let (kill-buffer-hook kill-buffer-query-functions
)
94 (kill-buffer buffer
)))
95 (let ((hook (assoc file after-load-alist
)))
97 (mapcar (function eval
) (cdr hook
))))
98 (unless (or nomessage noninteractive
)
100 (message "Loading %s (source)...done" file
)
101 (message "Loading %s...done" file
)))
104 ;; API (Application Program Interface) for charsets.
106 (defsubst charset-quoted-standard-p
(obj)
107 "Return t if OBJ is a quoted symbol, and is the name of a standard charset."
108 (and (listp obj
) (eq (car obj
) 'quote
)
109 (symbolp (car-safe (cdr obj
)))
110 (let ((vector (get (car-safe (cdr obj
)) 'charset
)))
111 (and (vectorp vector
)
112 (< (aref vector
0) 160)))))
114 (defsubst charsetp
(object)
115 "T if OBJECT is a charset."
116 (and (symbolp object
) (vectorp (get object
'charset
))))
118 (defsubst charset-info
(charset)
119 "Return a vector of information of CHARSET.
120 The elements of the vector are:
121 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
122 LEADING-CODE-BASE, LEADING-CODE-EXT,
123 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
124 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
127 CHARSET-ID (integer) is the identification number of the charset.
128 BYTES (integer) is the length of multi-byte form of a character in
129 the charset: one of 1, 2, 3, and 4.
130 DIMENSION (integer) is the number of bytes to represent a character of
132 CHARS (integer) is the number of characters in a dimension: 94 or 96.
133 WIDTH (integer) is the number of columns a character in the charset
134 occupies on the screen: one of 0, 1, and 2.
135 DIRECTION (integer) is the rendering direction of characters in the
136 charset when rendering. If 0, render from left to right, else
137 render from right to left.
138 LEADING-CODE-BASE (integer) is the base leading-code for the
140 LEADING-CODE-EXT (integer) is the extended leading-code for the
141 charset. All charsets of less than 0xA0 has the value 0.
142 ISO-FINAL-CHAR (character) is the final character of the
143 corresponding ISO 2022 charset. If the charset is not assigned
144 any final character, the value is -1.
145 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
146 while encoding to variants of ISO 2022 coding system, one of the
147 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
148 If the charset is not assigned any final character, the value is -1.
149 REVERSE-CHARSET (integer) is the charset which differs only in
150 LEFT-TO-RIGHT value from the charset. If there's no such a
151 charset, the value is -1.
152 SHORT-NAME (string) is the short name to refer to the charset.
153 LONG-NAME (string) is the long name to refer to the charset
154 DESCRIPTION (string) is the description string of the charset.
155 PLIST (property list) may contain any type of information a user
156 want to put and get by functions `put-charset-property' and
157 `get-charset-property' respectively."
158 (get charset
'charset
))
160 ;; It is better not to use backquote in this file,
161 ;; because that makes a bootstrapping problem
162 ;; if you need to recompile all the Lisp files using interpreted code.
164 (defmacro charset-id
(charset)
165 "Return charset identification number of CHARSET."
166 (if (charset-quoted-standard-p charset
)
167 (aref (charset-info (nth 1 charset
)) 0)
168 (list 'aref
(list 'charset-info charset
) 0)))
170 (defmacro charset-bytes
(charset)
171 "Return bytes of CHARSET.
172 See the function `charset-info' for more detail."
173 (if (charset-quoted-standard-p charset
)
174 (aref (charset-info (nth 1 charset
)) 1)
175 (list 'aref
(list 'charset-info charset
) 1)))
177 (defmacro charset-dimension
(charset)
178 "Return dimension of CHARSET.
179 See the function `charset-info' for more detail."
180 (if (charset-quoted-standard-p charset
)
181 (aref (charset-info (nth 1 charset
)) 2)
182 (list 'aref
(list 'charset-info charset
) 2)))
184 (defmacro charset-chars
(charset)
185 "Return character numbers contained in a dimension of CHARSET.
186 See the function `charset-info' for more detail."
187 (if (charset-quoted-standard-p charset
)
188 (aref (charset-info (nth 1 charset
)) 3)
189 (list 'aref
(list 'charset-info charset
) 3)))
191 (defmacro charset-width
(charset)
192 "Return width (how many column occupied on a screen) of CHARSET.
193 See the function `charset-info' for more detail."
194 (if (charset-quoted-standard-p charset
)
195 (aref (charset-info (nth 1 charset
)) 4)
196 (list 'aref
(list 'charset-info charset
) 4)))
198 (defmacro charset-direction
(charset)
199 "Return direction of CHARSET.
200 See the function `charset-info' for more detail."
201 (if (charset-quoted-standard-p charset
)
202 (aref (charset-info (nth 1 charset
)) 5)
203 (list 'aref
(list 'charset-info charset
) 5)))
205 (defmacro charset-iso-final-char
(charset)
206 "Return final char of CHARSET.
207 See the function `charset-info' for more detail."
208 (if (charset-quoted-standard-p charset
)
209 (aref (charset-info (nth 1 charset
)) 8)
210 (list 'aref
(list 'charset-info charset
) 8)))
212 (defmacro charset-iso-graphic-plane
(charset)
213 "Return graphic plane of CHARSET.
214 See the function `charset-info' for more detail."
215 (if (charset-quoted-standard-p charset
)
216 (aref (charset-info (nth 1 charset
)) 9)
217 (list 'aref
(list 'charset-info charset
) 9)))
219 (defmacro charset-reverse-charset
(charset)
220 "Return reverse charset of CHARSET.
221 See the function `charset-info' for more detail."
222 (if (charset-quoted-standard-p charset
)
223 (aref (charset-info (nth 1 charset
)) 10)
224 (list 'aref
(list 'charset-info charset
) 10)))
226 (defmacro charset-short-name
(charset)
227 "Return short name of CHARSET.
228 See the function `charset-info' for more detail."
229 (if (charset-quoted-standard-p charset
)
230 (aref (charset-info (nth 1 charset
)) 11)
231 (list 'aref
(list 'charset-info charset
) 11)))
233 (defmacro charset-long-name
(charset)
234 "Return long name of CHARSET.
235 See the function `charset-info' for more detail."
236 (if (charset-quoted-standard-p charset
)
237 (aref (charset-info (nth 1 charset
)) 12)
238 (list 'aref
(list 'charset-info charset
) 12)))
240 (defmacro charset-description
(charset)
241 "Return description of CHARSET.
242 See the function `charset-info' for more detail."
243 (if (charset-quoted-standard-p charset
)
244 (aref (charset-info (nth 1 charset
)) 13)
245 (list 'aref
(list 'charset-info charset
) 13)))
247 (defmacro charset-plist
(charset)
248 "Return list charset property of CHARSET.
249 See the function `charset-info' for more detail."
251 (if (charset-quoted-standard-p charset
)
252 (charset-info (nth 1 charset
))
253 (list 'charset-info charset
))
256 (defun set-charset-plist (charset plist
)
257 "Set CHARSET's property list to PLIST, and return PLIST."
258 (aset (charset-info charset
) 14 plist
))
260 (defun make-char (charset &optional code1 code2
)
261 "Return a character of CHARSET whose position codes are CODE1 and CODE2.
262 CODE1 and CODE2 are optional, but if you don't supply
263 sufficient position codes, return a generic character which stands for
264 all characters or group of characters in the character set.
265 A generic character can be used to index a char table (e.g. syntax-table).
267 Such character sets as ascii, eight-bit-control, and eight-bit-graphic
268 don't have corresponding generic characters. If CHARSET is one of
269 them and you don't supply CODE1, return the character of the smallest
272 If CODE1 or CODE2 are invalid (out of range), this function signals an
273 error. However, the eighth bit of both CODE1 and CODE2 is zeroed
274 before they are used to index CHARSET. Thus you may use, say, the
275 actual ISO 8859 character code rather than subtracting 128, as you
276 would need to index the corresponding Emacs charset."
277 (make-char-internal (charset-id charset
) code1 code2
))
279 (put 'make-char
'byte-compile
282 (let ((charset (nth 1 form
)))
283 (if (charset-quoted-standard-p charset
)
284 (byte-compile-normal-call
285 (cons 'make-char-internal
286 (cons (charset-id (nth 1 charset
)) (nthcdr 2 form
))))
287 (byte-compile-normal-call
288 (cons 'make-char-internal
289 (cons (list 'charset-id charset
) (nthcdr 2 form
)))))))))
291 (defun charset-list ()
292 "Return list of charsets ever defined.
294 This function is provided for backward compatibility.
295 Now we have the variable `charset-list'."
298 (defsubst generic-char-p
(char)
299 "Return t if and only if CHAR is a generic character.
300 See also the documentation of `make-char'."
302 (let ((l (split-char char
)))
303 (and (or (= (nth 1 l
) 0) (eq (nth 2 l
) 0))
304 (not (eq (car l
) 'composition
))))))
306 (defun decode-char (ccs code-point
&optional restriction
)
307 "Return character specified by coded character set CCS and CODE-POINT in it.
308 Return nil if such a character is not supported.
309 Currently the only supported coded character set is `ucs' (ISO/IEC
310 10646: Universal Multi-Octet Coded Character Set), and the result is
311 translated through the translation-table named
312 `utf-translation-table-for-decode' or the translation-hash-table named
313 `utf-subst-table-for-decode'.
315 Optional argument RESTRICTION specifies a way to map the pair of CCS
316 and CODE-POINT to a character. Currently not supported and just ignored."
319 (or (gethash code-point
320 (get 'utf-subst-table-for-decode
'translation-hash-table
))
325 (make-char 'latin-iso8859-1 code-point
))
326 ((< code-point
#x2500
)
327 (setq code-point
(- code-point
#x0100
))
328 (make-char 'mule-unicode-0100-24ff
329 (+ (/ code-point
96) 32) (+ (% code-point
96) 32)))
330 ((< code-point
#x3400
)
331 (setq code-point
(- code-point
#x2500
))
332 (make-char 'mule-unicode-2500-33ff
333 (+ (/ code-point
96) 32) (+ (% code-point
96) 32)))
334 ((and (>= code-point
#xe000
) (< code-point
#x10000
))
335 (setq code-point
(- code-point
#xe000
))
336 (make-char 'mule-unicode-e000-ffff
337 (+ (/ code-point
96) 32)
338 (+ (% code-point
96) 32))))))
340 (or (aref (get 'utf-translation-table-for-decode
341 'translation-table
) c
)
344 (defun encode-char (char ccs
&optional restriction
)
345 "Return code-point in coded character set CCS that corresponds to CHAR.
346 Return nil if CHAR is not included in CCS.
347 Currently the only supported coded character set is `ucs' (ISO/IEC
348 10646: Universal Multi-Octet Coded Character Set), and CHAR is first
349 translated through the translation-table named
350 `utf-translation-table-for-encode' or the translation-hash-table named
351 `utf-subst-table-for-encode'.
353 CHAR should be in one of these charsets:
354 ascii, latin-iso8859-1, mule-unicode-0100-24ff, mule-unicode-2500-33ff,
355 mule-unicode-e000-ffff, eight-bit-control
356 Otherwise, return nil.
358 Optional argument RESTRICTION specifies a way to map CHAR to a
359 code-point in CCS. Currently not supported and just ignored."
360 (let* ((split (split-char char
))
361 (charset (car split
))
364 (or (gethash char
(get 'utf-subst-table-for-encode
365 'translation-hash-table
))
366 (let ((table (get 'utf-translation-table-for-encode
367 'translation-table
)))
368 (setq trans
(aref table char
))
370 (setq split
(split-char trans
)
371 charset
(car split
)))
372 (cond ((eq charset
'ascii
)
374 ((eq charset
'latin-iso8859-1
)
375 (+ (nth 1 split
) 128))
376 ((eq charset
'mule-unicode-0100-24ff
)
377 (+ #x0100
(+ (* (- (nth 1 split
) 32) 96)
378 (- (nth 2 split
) 32))))
379 ((eq charset
'mule-unicode-2500-33ff
)
380 (+ #x2500
(+ (* (- (nth 1 split
) 32) 96)
381 (- (nth 2 split
) 32))))
382 ((eq charset
'mule-unicode-e000-ffff
)
383 (+ #xe000
(+ (* (- (nth 1 split
) 32) 96)
384 (- (nth 2 split
) 32))))
385 ((eq charset
'eight-bit-control
)
389 ;; Coding system stuff
391 ;; Coding system is a symbol that has the property `coding-system'.
393 ;; The value of the property `coding-system' is a vector of the
395 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
396 ;; We call this vector as coding-spec. See comments in src/coding.c
399 (defconst coding-spec-type-idx
0)
400 (defconst coding-spec-mnemonic-idx
1)
401 (defconst coding-spec-doc-string-idx
2)
402 (defconst coding-spec-plist-idx
3)
403 (defconst coding-spec-flags-idx
4)
405 ;; PLIST is a property list of a coding system. To share PLIST among
406 ;; alias coding systems, a coding system has PLIST in coding-spec
407 ;; instead of having it in normal property list of Lisp symbol.
408 ;; Here's a list of coding system properties currently being used.
412 ;; The value is a coding category the coding system belongs to. The
413 ;; function `make-coding-system' sets this value automatically
414 ;; unless its argument PROPERTIES specifies this property.
416 ;; o alias-coding-systems
418 ;; The value is a list of coding systems of the same alias group. The
419 ;; first element is the coding system made at first, which we call as
420 ;; `base coding system'. The function `make-coding-system' sets this
421 ;; value automatically and `define-coding-system-alias' updates it.
423 ;; See the documentation of make-coding-system for the meanings of the
424 ;; following properties.
426 ;; o post-read-conversion
427 ;; o pre-write-conversion
428 ;; o translation-table-for-decode
429 ;; o translation-table-for-encode
433 ;; o valid-codes (meaningful only for a coding system based on CCL)
436 (defsubst coding-system-spec
(coding-system)
437 "Return coding-spec of CODING-SYSTEM."
438 (get (check-coding-system coding-system
) 'coding-system
))
440 (defun coding-system-type (coding-system)
441 "Return the coding type of CODING-SYSTEM.
442 A coding type is an integer value indicating the encoding method
443 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
444 (aref (coding-system-spec coding-system
) coding-spec-type-idx
))
446 (defun coding-system-mnemonic (coding-system)
447 "Return the mnemonic character of CODING-SYSTEM.
448 The mnemonic character of a coding system is used in mode line
449 to indicate the coding system. If the arg is nil, return ?-."
450 (let ((spec (coding-system-spec coding-system
)))
451 (if spec
(aref spec coding-spec-mnemonic-idx
) ?-
)))
453 (defun coding-system-doc-string (coding-system)
454 "Return the documentation string for CODING-SYSTEM."
455 (aref (coding-system-spec coding-system
) coding-spec-doc-string-idx
))
457 (defun coding-system-plist (coding-system)
458 "Return the property list of CODING-SYSTEM."
459 (aref (coding-system-spec coding-system
) coding-spec-plist-idx
))
461 (defun coding-system-flags (coding-system)
462 "Return `flags' of CODING-SYSTEM.
463 A `flags' of a coding system is a vector of length 32 indicating detailed
464 information of a coding system. See the function `make-coding-system'
466 (aref (coding-system-spec coding-system
) coding-spec-flags-idx
))
468 (defun coding-system-get (coding-system prop
)
469 "Extract a value from CODING-SYSTEM's property list for property PROP."
470 (plist-get (coding-system-plist coding-system
) prop
))
472 (defun coding-system-put (coding-system prop val
)
473 "Change value in CODING-SYSTEM's property list PROP to VAL."
474 (let ((plist (coding-system-plist coding-system
)))
476 (plist-put plist prop val
)
477 (aset (coding-system-spec coding-system
) coding-spec-plist-idx
480 (defun coding-system-category (coding-system)
481 "Return the coding category of CODING-SYSTEM.
482 See also `coding-category-list'."
483 (coding-system-get coding-system
'coding-category
))
485 (defun coding-system-base (coding-system)
486 "Return the base coding system of CODING-SYSTEM.
487 A base coding system is what made by `make-coding-system'.
488 Any alias nor subsidiary coding systems are not base coding system."
489 (car (coding-system-get coding-system
'alias-coding-systems
)))
491 (defalias 'coding-system-parent
'coding-system-base
)
492 (make-obsolete 'coding-system-parent
'coding-system-base
"20.3")
494 ;; Coding system also has a property `eol-type'.
496 ;; This property indicates how the coding system handles end-of-line
497 ;; format. The value is integer 0, 1, 2, or a vector of three coding
498 ;; systems. Each integer value 0, 1, and 2 indicates the format of
499 ;; end-of-line LF, CRLF, and CR respectively. A vector value
500 ;; indicates that the format of end-of-line should be detected
501 ;; automatically. Nth element of the vector is the subsidiary coding
502 ;; system whose `eol-type' property is N.
504 (defun coding-system-eol-type (coding-system)
505 "Return eol-type of CODING-SYSTEM.
506 An eol-type is integer 0, 1, 2, or a vector of coding systems.
508 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
509 CRLF, and CR respectively.
511 A vector value indicates that a format of end-of-line should be
512 detected automatically. Nth element of the vector is the subsidiary
513 coding system whose eol-type is N."
514 (get coding-system
'eol-type
))
516 (defun coding-system-eol-type-mnemonic (coding-system)
517 "Return the string indicating end-of-line format of CODING-SYSTEM."
518 (let* ((eol-type (coding-system-eol-type coding-system
))
519 (val (cond ((eq eol-type
0) eol-mnemonic-unix
)
520 ((eq eol-type
1) eol-mnemonic-dos
)
521 ((eq eol-type
2) eol-mnemonic-mac
)
522 (t eol-mnemonic-undecided
))))
525 (char-to-string val
))))
527 (defun coding-system-lessp (x y
)
528 (cond ((eq x
'no-conversion
) t
)
529 ((eq y
'no-conversion
) nil
)
530 ((eq x
'emacs-mule
) t
)
531 ((eq y
'emacs-mule
) nil
)
532 ((eq x
'undecided
) t
)
533 ((eq y
'undecided
) nil
)
534 (t (let ((c1 (coding-system-mnemonic x
))
535 (c2 (coding-system-mnemonic y
)))
536 (or (< (downcase c1
) (downcase c2
))
537 (and (not (> (downcase c1
) (downcase c2
)))
540 (defun add-to-coding-system-list (coding-system)
541 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
542 (if (or (null coding-system-list
)
543 (coding-system-lessp coding-system
(car coding-system-list
)))
544 (setq coding-system-list
(cons coding-system coding-system-list
))
545 (let ((len (length coding-system-list
))
546 mid
(tem coding-system-list
))
548 (setq mid
(nthcdr (/ len
2) tem
))
549 (if (coding-system-lessp (car mid
) coding-system
)
551 len
(- len
(/ len
2)))
552 (setq len
(/ len
2))))
553 (setcdr tem
(cons coding-system
(cdr tem
))))))
555 (defun coding-system-list (&optional base-only
)
556 "Return a list of all existing non-subsidiary coding systems.
557 If optional arg BASE-ONLY is non-nil, only base coding systems are listed.
558 The value doesn't include subsidiary coding systems which are what
559 made from bases and aliases automatically for various end-of-line
560 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
561 (let* ((codings (copy-sequence coding-system-list
))
562 (tail (cons nil codings
)))
563 ;; Remove subsidiary coding systems (eol variants) and alias
564 ;; coding systems (if necessary).
566 (let* ((coding (car (cdr tail
)))
567 (aliases (coding-system-get coding
'alias-coding-systems
)))
569 ;; CODING is an eol variant if not in ALIASES.
570 (not (memq coding aliases
))
571 ;; CODING is an alias if it is not car of ALIASES.
572 (and base-only
(not (eq coding
(car aliases
)))))
573 (setcdr tail
(cdr (cdr tail
)))
574 (setq tail
(cdr tail
)))))
577 (defun map-charset-chars (func charset
)
578 "Use FUNC to map over all characters in CHARSET for side effects.
579 FUNC is a function of two args, the start and end (inclusive) of a
580 character code range. Thus FUNC should iterate over [START, END]."
581 (let* ((dim (charset-dimension charset
))
582 (chars (charset-chars charset
))
583 (start (if (= chars
94)
588 (make-char charset start
)
589 (make-char charset
(+ start chars -
1)))
592 (make-char charset
(+ i start
) start
)
593 (make-char charset
(+ i start
) (+ start chars -
1)))))))
595 (defun register-char-codings (coding-system safe-chars
)
596 "This is an obsolete function.
597 It exists just for backward compatibility, and it does nothing.")
598 (make-obsolete 'register-char-codings
599 "Unnecessary function. Calling it has no effect."
602 (defconst char-coding-system-table nil
603 "This is an obsolete variable.
604 It exists just for backward compatibility, and the value is always nil.")
606 (defun make-subsidiary-coding-system (coding-system)
607 "Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM."
608 (let ((coding-spec (coding-system-spec coding-system
))
609 (subsidiaries (vector (intern (format "%s-unix" coding-system
))
610 (intern (format "%s-dos" coding-system
))
611 (intern (format "%s-mac" coding-system
))))
615 (put (aref subsidiaries i
) 'coding-system coding-spec
)
616 (put (aref subsidiaries i
) 'eol-type i
)
617 (add-to-coding-system-list (aref subsidiaries i
))
618 (setq coding-system-alist
619 (cons (list (symbol-name (aref subsidiaries i
)))
620 coding-system-alist
))
624 (defun transform-make-coding-system-args (name type
&optional doc-string props
)
625 "For internal use only.
626 Transform XEmacs style args for `make-coding-system' to Emacs style.
627 Value is a list of transformed arguments."
628 (let ((mnemonic (string-to-char (or (plist-get props
'mnemonic
) "?")))
629 (eol-type (plist-get props
'eol-type
))
632 ((eq eol-type
'lf
) (setq eol-type
'unix
))
633 ((eq eol-type
'crlf
) (setq eol-type
'dos
))
634 ((eq eol-type
'cr
) (setq eol-type
'mac
)))
635 (if (setq tmp
(plist-get props
'post-read-conversion
))
636 (setq properties
(plist-put properties
'post-read-conversion tmp
)))
637 (if (setq tmp
(plist-get props
'pre-write-conversion
))
638 (setq properties
(plist-put properties
'pre-write-conversion tmp
)))
640 ((eq type
'shift-jis
)
641 `(,name
1 ,mnemonic
,doc-string
() ,properties
,eol-type
))
642 ((eq type
'iso2022
) ; This is not perfect.
643 (if (plist-get props
'escape-quoted
)
644 (error "escape-quoted is not supported: %S"
645 `(,name
,type
,doc-string
,props
)))
646 (let ((g0 (plist-get props
'charset-g0
))
647 (g1 (plist-get props
'charset-g1
))
648 (g2 (plist-get props
'charset-g2
))
649 (g3 (plist-get props
'charset-g3
))
652 (eq (cadr (assoc 'latin-jisx0201
653 (plist-get props
'input-charset-conversion
)))
655 (eq (cadr (assoc 'ascii
656 (plist-get props
'output-charset-conversion
)))
660 (eq (cadr (assoc 'japanese-jisx0208-1978
661 (plist-get props
'input-charset-conversion
)))
663 (eq (cadr (assoc 'japanese-jisx0208
664 (plist-get props
'output-charset-conversion
)))
665 'japanese-jisx0208-1978
))))
667 (if (plist-get props
'force-g0-on-output
)
671 (if (plist-get props
'force-g1-on-output
)
675 (if (plist-get props
'force-g2-on-output
)
679 (if (plist-get props
'force-g3-on-output
)
682 `(,name
2 ,mnemonic
,doc-string
684 ,(plist-get props
'short
)
685 ,(not (plist-get props
'no-ascii-eol
))
686 ,(not (plist-get props
'no-ascii-cntl
))
687 ,(plist-get props
'seven
)
689 ,(not (plist-get props
'lock-shift
))
692 ,(plist-get props
'no-iso6429
)
694 ,properties
,eol-type
)))
696 `(,name
3 ,mnemonic
,doc-string
() ,properties
,eol-type
))
698 `(,name
4 ,mnemonic
,doc-string
699 (,(plist-get props
'decode
) .
,(plist-get props
'encode
))
700 ,properties
,eol-type
))
702 (error "unsupported XEmacs style make-coding-style arguments: %S"
703 `(,name
,type
,doc-string
,props
))))))
705 (defun make-coding-system (coding-system type mnemonic doc-string
710 "Define a new coding system CODING-SYSTEM (symbol).
711 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
712 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
713 in the following format:
714 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
716 TYPE is an integer value indicating the type of the coding system as follows:
717 0: Emacs internal format,
718 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PCs,
719 2: ISO-2022 including many variants,
720 3: Big5 used mainly on Chinese PCs,
721 4: private, CCL programs provide encoding/decoding algorithm,
722 5: Raw-text, which means that text contains random 8-bit codes.
724 MNEMONIC is a character to be displayed on mode line for the coding system.
726 DOC-STRING is a documentation string for the coding system.
728 FLAGS specifies more detailed information of the coding system as follows:
730 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
731 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
732 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
733 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
734 SAFE, ACCEPT-LATIN-EXTRA-CODE.
735 CHARSETn are character sets initially designated to Gn graphic registers.
736 If CHARSETn is nil, Gn is never used.
737 If CHARSETn is t, Gn can be used but nothing designated initially.
738 If CHARSETn is a list of character sets, those character sets are
739 designated to Gn on output, but nothing designated to Gn initially.
740 But, character set `ascii' can be designated only to G0.
741 SHORT-FORM non-nil means use short designation sequence on output.
742 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
743 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
745 SEVEN non-nil means use 7-bit code only on output.
746 LOCKING-SHIFT non-nil means use locking-shift.
747 SINGLE-SHIFT non-nil means use single-shift.
748 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
749 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
750 NO-ISO6429 non-nil means not use ISO6429's direction specification.
751 INIT-BOL non-nil means any designation state is assumed to be reset
752 to initial at each beginning of line on output.
753 DESIGNATION-BOL non-nil means designation sequences should be placed
754 at beginning of line on output.
755 SAFE non-nil means convert unsafe characters to `?' on output.
756 Characters not specified in the property `safe-charsets' nor
757 `safe-chars' are unsafe.
758 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
759 a code specified in `latin-extra-code-table' (which see) as a valid
760 code of the coding system.
762 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
763 decoding and encoding. CCL programs should be specified by their
766 PROPERTIES is an alist of properties vs the corresponding values. The
767 following properties are recognized:
769 o post-read-conversion
771 The value is a function to call after some text is inserted and
772 decoded by the coding system itself and before any functions in
773 `after-insert-functions' are called. The argument of this
774 function is the same as for a function in
775 `after-insert-file-functions', i.e. LENGTH of the text inserted,
776 with point at the head of the text to be decoded.
778 o pre-write-conversion
780 The value is a function to call after all functions in
781 `write-region-annotate-functions' and `buffer-file-format' are
782 called, and before the text is encoded by the coding system itself.
783 The arguments to this function are the same as those of a function
784 in `write-region-annotate-functions', i.e. FROM and TO, specifying
787 o translation-table-for-decode
789 The value is a translation table to be applied on decoding. See
790 the function `make-translation-table' for the format of translation
791 table. This is not applicable to type 4 (CCL-based) coding systems.
793 o translation-table-for-encode
795 The value is a translation table to be applied on encoding. This is
796 not applicable to type 4 (CCL-based) coding systems.
800 The value is a char table. If a character has non-nil value in it,
801 the character is safely supported by the coding system. This
802 overrides the specification of safe-charsets.
806 The value is a list of charsets safely supported by the coding
807 system. The value t means that all charsets Emacs handles are
808 supported. Even if some charset is not in this list, it doesn't
809 mean that the charset can't be encoded in the coding system;
810 it just means that some other receiver of text encoded
811 in the coding system won't be able to handle that charset.
815 The value is a symbol whose name is the `MIME-charset' parameter of
818 o valid-codes (meaningful only for a coding system based on CCL)
820 The value is a list to indicate valid byte ranges of the encoded
821 file. Each element of the list is an integer or a cons of integer.
822 In the former case, the integer value is a valid byte code. In the
823 latter case, the integers specify the range of valid byte codes.
825 o composition (meaningful only when TYPE is 0 or 2)
827 If the value is non-nil, the coding system preserves composition
830 These properties are set in PLIST, a property list. This function
831 also sets properties `coding-category' and `alias-coding-systems'
834 EOL-TYPE specifies the EOL type of the coding-system in one of the
837 o symbol (unix, dos, or mac)
839 The symbol `unix' means Unix-like EOL (LF), `dos' means
840 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
842 o number (0, 1, or 2)
844 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
847 o vector of coding-systems of length 3
849 The EOL type is detected automatically for the coding system.
850 And, according to the detected EOL type, one of the coding
851 systems in the vector is selected. Elements of the vector
852 corresponds to Unix-like EOL, DOS-like EOL, and Mac-like EOL
855 Kludgy features for backward compatibility:
857 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
858 treated as a compiled CCL code.
860 2. If PROPERTIES is just a list of character sets, the list is set as
861 a value of `safe-charsets' in PLIST."
863 ;; For compatiblity with XEmacs, we check the type of TYPE. If it
864 ;; is a symbol, perhaps, this function is called with XEmacs-style
865 ;; arguments. Here, try to transform that kind of arguments to
868 (let ((args (transform-make-coding-system-args coding-system type
869 mnemonic doc-string
)))
870 (setq coding-system
(car args
)
872 mnemonic
(nth 2 args
)
873 doc-string
(nth 3 args
)
875 properties
(nth 5 args
)
876 eol-type
(nth 6 args
))))
878 ;; Set a value of `coding-system' property.
879 (let ((coding-spec (make-vector 5 nil
))
880 (no-initial-designation t
)
881 (no-alternative-designation t
)
882 (accept-latin-extra-code nil
)
884 (if (or (not (integerp type
)) (< type
0) (> type
5))
885 (error "TYPE argument must be 0..5"))
886 (if (or (not (integerp mnemonic
)) (<= mnemonic ?
) (> mnemonic
127))
887 (error "MNEMONIC argument must be an ASCII printable character"))
888 (aset coding-spec coding-spec-type-idx type
)
889 (aset coding-spec coding-spec-mnemonic-idx mnemonic
)
890 (aset coding-spec coding-spec-doc-string-idx
891 (purecopy (if (stringp doc-string
) doc-string
"")))
893 (setq coding-category
'coding-category-emacs-mule
))
895 (setq coding-category
'coding-category-sjis
))
896 ((= type
2) ; ISO2022
898 (vec (make-vector 32 nil
))
902 (let ((charset (car fl
)))
903 (if (and no-initial-designation
905 (or (charsetp charset
)
907 (charsetp (car charset
)))))
908 (setq no-initial-designation nil
))
909 (if (charsetp charset
)
910 (if (= i
1) (setq g1-designation charset
))
915 (setq elt
(car tail
))
917 (setq no-alternative-designation nil
)
918 (if (and elt
(not (charsetp elt
)))
919 (error "Invalid charset: %s" elt
)))
920 (setq tail
(cdr tail
)))
921 (setq g1-designation
(car charset
)))
924 (setq no-alternative-designation nil
)
925 (error "Invalid charset: %s" charset
)))))
926 (aset vec i charset
))
927 (setq fl
(cdr fl
) i
(1+ i
)))
928 (while (and (< i
32) fl
)
929 (aset vec i
(car fl
))
930 (if (and (= i
16) ; ACCEPT-LATIN-EXTRA-CODE
932 (setq accept-latin-extra-code t
))
933 (setq fl
(cdr fl
) i
(1+ i
)))
934 (aset coding-spec
4 vec
)
935 (setq coding-category
936 (if (aref vec
8) ; Use locking-shift.
937 (or (and (aref vec
7) 'coding-category-iso-7-else
)
938 'coding-category-iso-8-else
)
939 (if (aref vec
7) ; 7-bit only.
940 (if (aref vec
9) ; Use single-shift.
941 'coding-category-iso-7-else
942 (if no-alternative-designation
943 'coding-category-iso-7-tight
944 'coding-category-iso-7
))
945 (if (or no-initial-designation
946 (not no-alternative-designation
))
947 'coding-category-iso-8-else
948 (if (and (charsetp g1-designation
)
949 (= (charset-dimension g1-designation
) 2))
950 'coding-category-iso-8-2
951 'coding-category-iso-8-1
)))))))
953 (setq coding-category
'coding-category-big5
))
954 ((= type
4) ; private
955 (setq coding-category
'coding-category-ccl
)
956 (if (not (consp flags
))
957 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
958 (let ((decoder (check-ccl-program
960 (intern (format "%s-decoder" coding-system
))))
961 (encoder (check-ccl-program
963 (intern (format "%s-encoder" coding-system
)))))
964 (if (and decoder encoder
)
965 (aset coding-spec
4 (cons decoder encoder
))
966 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
968 (setq coding-category
'coding-category-raw-text
)))
970 (let ((plist (list 'coding-category coding-category
971 'alias-coding-systems
(list coding-system
))))
972 (if no-initial-designation
973 (plist-put plist
'no-initial-designation t
))
975 (or (eq properties t
)
976 (not (consp (car properties
)))))
977 ;; In the old version, the arg PROPERTIES is a list to be
978 ;; set in PLIST as a value of property `safe-charsets'.
979 (setq properties
(list (cons 'safe-charsets properties
))))
980 ;; In the current version PROPERTIES is a property list.
981 ;; Reflect it into PLIST one by one while handling safe-chars
983 (let ((safe-charsets (cdr (assq 'safe-charsets properties
)))
984 (safe-chars (cdr (assq 'safe-chars properties
)))
987 ;; If only safe-charsets is specified, make a char-table from
988 ;; it, and store that char-table as the value of `safe-chars'.
989 (if (and (not safe-chars
) safe-charsets
)
991 (if (eq safe-charsets t
)
993 (setq safe-chars
(make-char-table 'safe-chars
))
995 (setq charset
(car safe-charsets
)
996 safe-charsets
(cdr safe-charsets
))
997 (cond ((eq charset
'ascii
)) ; just ignore
998 ((eq charset
'eight-bit-control
)
1001 (aset safe-chars i t
)
1003 ((eq charset
'eight-bit-graphic
)
1006 (aset safe-chars i t
)
1009 (aset safe-chars
(make-char charset
) t
))))
1010 (if accept-latin-extra-code
1013 (if (aref latin-extra-code-table i
)
1014 (aset safe-chars i t
))
1016 (setq l
(cons (cons 'safe-chars safe-chars
) l
))))
1018 (setq prop
(car (car l
)) val
(cdr (car l
)) l
(cdr l
))
1019 (if (eq prop
'safe-chars
)
1021 (if (and (symbolp val
)
1022 (get val
'translation-table
))
1023 (setq safe-chars
(get val
'translation-table
)))
1024 (setq val safe-chars
)))
1025 (plist-put plist prop val
)))
1026 ;; The property `coding-category' may have been set differently
1027 ;; through PROPERTIES.
1028 (setq coding-category
(plist-get plist
'coding-category
))
1029 (aset coding-spec coding-spec-plist-idx plist
))
1030 (put coding-system
'coding-system coding-spec
)
1031 (put coding-category
'coding-systems
1032 (cons coding-system
(get coding-category
'coding-systems
))))
1034 ;; Next, set a value of `eol-type' property.
1036 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
1037 ;; systems, each corresponds to a coding system for the detected
1039 (setq eol-type
(make-subsidiary-coding-system coding-system
)))
1041 (cond ((or (eq eol-type
'unix
) (null eol-type
))
1047 ((or (and (vectorp eol-type
)
1048 (= (length eol-type
) 3))
1049 (and (numberp eol-type
)
1050 (and (>= eol-type
0)
1054 (error "Invalid EOL-TYPE spec:%S" eol-type
))))
1055 (put coding-system
'eol-type eol-type
)
1057 (define-coding-system-internal coding-system
)
1059 ;; At last, register CODING-SYSTEM in `coding-system-list' and
1060 ;; `coding-system-alist'.
1061 (add-to-coding-system-list coding-system
)
1062 (setq coding-system-alist
(cons (list (symbol-name coding-system
))
1063 coding-system-alist
))
1065 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
1066 ;; XXX-with-esc variants.
1067 (let ((coding-category (coding-system-category coding-system
)))
1068 (if (or (eq coding-category
'coding-category-iso-8-1
)
1069 (eq coding-category
'coding-category-iso-8-2
))
1070 (let ((esc (intern (concat (symbol-name coding-system
) "-with-esc")))
1071 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system
))
1072 (safe-charsets (assq 'safe-charsets properties
))
1073 (mime-charset (assq 'mime-charset properties
)))
1075 (setcdr safe-charsets t
)
1076 (setq properties
(cons (cons 'safe-charsets t
) properties
)))
1078 (setcdr mime-charset nil
))
1079 (make-coding-system esc type mnemonic doc
1080 (if (listp (car flags
))
1081 (cons (append (car flags
) '(t)) (cdr flags
))
1082 (cons (list (car flags
) t
) (cdr flags
)))
1087 (put 'safe-chars
'char-table-extra-slots
0)
1089 (defun define-coding-system-alias (alias coding-system
)
1090 "Define ALIAS as an alias for coding system CODING-SYSTEM."
1091 (put alias
'coding-system
(coding-system-spec coding-system
))
1092 (add-to-coding-system-list alias
)
1093 (setq coding-system-alist
(cons (list (symbol-name alias
))
1094 coding-system-alist
))
1095 (let ((eol-type (coding-system-eol-type coding-system
)))
1096 (if (vectorp eol-type
)
1098 (nconc (coding-system-get alias
'alias-coding-systems
) (list alias
))
1099 (put alias
'eol-type
(make-subsidiary-coding-system alias
)))
1100 (put alias
'eol-type eol-type
))))
1102 (defun merge-coding-systems (first second
)
1103 "Fill in any unspecified aspects of coding system FIRST from SECOND.
1104 Return the resulting coding system."
1105 (let ((base (coding-system-base second
))
1106 (eol (coding-system-eol-type second
)))
1107 ;; If FIRST doesn't specify text conversion, merge with that of SECOND.
1108 (if (eq (coding-system-base first
) 'undecided
)
1109 (setq first
(coding-system-change-text-conversion first base
)))
1110 ;; If FIRST doesn't specify eol conversion, merge with that of SECOND.
1111 (if (and (vectorp (coding-system-eol-type first
))
1112 (numberp eol
) (>= eol
0) (<= eol
2))
1113 (setq first
(coding-system-change-eol-conversion
1117 (defun set-buffer-file-coding-system (coding-system &optional force
)
1118 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1119 This means that when you save the buffer, it will be converted
1120 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
1121 use \\[list-coding-systems].
1123 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1124 leaves the end-of-line conversion unspecified, FORCE controls what to
1125 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1126 buffer's previous `buffer-file-coding-system' value (if it is
1127 specified there). Otherwise, levae it unspecified.
1129 This marks the buffer modified so that the succeeding \\[save-buffer]
1130 surely saves the buffer with CODING-SYSTEM. From a program, if you
1131 don't want to mark the buffer modified, just set the variable
1132 `buffer-file-coding-system' directly."
1133 (interactive "zCoding system for saving file (default, nil): \nP")
1134 (check-coding-system coding-system
)
1135 (if (and coding-system buffer-file-coding-system
(null force
))
1137 (merge-coding-systems coding-system buffer-file-coding-system
)))
1138 (setq buffer-file-coding-system coding-system
)
1139 ;; This is in case of an explicit call. Normally, `normal-mode' and
1140 ;; `set-buffer-major-mode-hook' take care of setting the table.
1141 (if (fboundp 'ucs-set-table-for-input
) ; don't lose when building
1142 (ucs-set-table-for-input))
1143 (set-buffer-modified-p t
)
1144 (force-mode-line-update))
1146 (defun revert-buffer-with-coding-system (coding-system &optional force
)
1147 "Visit the current buffer's file again using coding system CODING-SYSTEM.
1148 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1150 If CODING-SYSTEM leaves the text conversion unspecified, or if it
1151 leaves the end-of-line conversion unspecified, FORCE controls what to
1152 do. If FORCE is nil, get the unspecified aspect (or aspects) from the
1153 buffer's previous `buffer-file-coding-system' value (if it is
1154 specified there). Otherwise, determine it from the file contents as
1155 usual for visiting a file."
1156 (interactive "zCoding system for visited file (default, nil): \nP")
1157 (check-coding-system coding-system
)
1158 (if (and coding-system buffer-file-coding-system
(null force
))
1160 (merge-coding-systems coding-system buffer-file-coding-system
)))
1161 (let ((coding-system-for-read coding-system
))
1164 (defun set-file-name-coding-system (coding-system)
1165 "Set coding system for decoding and encoding file names to CODING-SYSTEM.
1166 It actually just set the variable `file-name-coding-system' (which
1167 see) to CODING-SYSTEM."
1168 (interactive "zCoding system for file names (default, nil): ")
1169 (check-coding-system coding-system
)
1170 (setq file-name-coding-system coding-system
))
1172 (defvar default-terminal-coding-system nil
1173 "Default value for the terminal coding system.
1174 This is normally set according to the selected language environment.
1175 See also the command `set-terminal-coding-system'.")
1177 (defun set-terminal-coding-system (coding-system)
1178 "Set coding system of your terminal to CODING-SYSTEM.
1179 All text output to the terminal will be encoded
1180 with the specified coding system.
1181 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1182 The default is determined by the selected language environment
1183 or by the previous use of this command."
1185 (list (let ((default (if (and (not (terminal-coding-system))
1186 default-terminal-coding-system
)
1187 default-terminal-coding-system
)))
1189 (format "Coding system for terminal display (default, %s): "
1192 (if (and (not coding-system
)
1193 (not (terminal-coding-system)))
1194 (setq coding-system default-terminal-coding-system
))
1196 (setq default-terminal-coding-system coding-system
))
1197 (set-terminal-coding-system-internal coding-system
)
1198 (redraw-frame (selected-frame)))
1200 (defvar default-keyboard-coding-system nil
1201 "Default value of the keyboard coding system.
1202 This is normally set according to the selected language environment.
1203 See also the command `set-keyboard-coding-system'.")
1205 (defun set-keyboard-coding-system (coding-system)
1206 "Set coding system for keyboard input to CODING-SYSTEM.
1207 In addition, this command enables Encoded-kbd minor mode.
1208 \(If CODING-SYSTEM is nil, Encoded-kbd mode is turned off -- see
1209 `encoded-kbd-mode'.)
1210 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1211 The default is determined by the selected language environment
1212 or by the previous use of this command."
1214 (list (let ((default (if (and (not (keyboard-coding-system))
1215 default-keyboard-coding-system
)
1216 default-keyboard-coding-system
)))
1218 (format "Coding system for keyboard input (default, %s): "
1221 (if (and (not coding-system
)
1222 (not (keyboard-coding-system)))
1223 (setq coding-system default-keyboard-coding-system
))
1225 (setq default-keyboard-coding-system coding-system
))
1226 (set-keyboard-coding-system-internal coding-system
)
1227 (setq keyboard-coding-system coding-system
)
1228 (encoded-kbd-mode (if coding-system
1 0)))
1230 (defcustom keyboard-coding-system nil
1231 "Specify coding system for keyboard input.
1232 If you set this on a terminal which can't distinguish Meta keys from
1233 8-bit characters, you will have to use ESC to type Meta characters.
1234 See Info node `Specify Coding' and Info node `Single-Byte Character Support'.
1236 On non-windowing terminals, this is set from the locale by default.
1238 Setting this variable directly does not take effect;
1239 use either M-x customize or \\[set-keyboard-coding-system]."
1240 :type
'(coding-system :tag
"Coding system")
1241 :link
'(info-link "(emacs)Specify Coding")
1242 :link
'(info-link "(emacs)Single-Byte Character Support")
1243 :set
(lambda (symbol value
)
1244 ;; Don't load encoded-kbd-mode unnecessarily.
1245 (if (or value
(boundp 'encoded-kbd-mode
))
1246 (set-keyboard-coding-system value
)
1247 (set-default 'keyboard-coding-system nil
))) ; must initialize
1252 (defun set-buffer-process-coding-system (decoding encoding
)
1253 "Set coding systems for the process associated with the current buffer.
1254 DECODING is the coding system to be used to decode input from the process,
1255 ENCODING is the coding system to be used to encode output to the process.
1257 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
1259 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1260 (let ((proc (get-buffer-process (current-buffer))))
1262 (error "No process")
1263 (check-coding-system decoding
)
1264 (check-coding-system encoding
)
1265 (set-process-coding-system proc decoding encoding
)))
1266 (force-mode-line-update))
1268 (defalias 'set-clipboard-coding-system
'set-selection-coding-system
)
1270 (defun set-selection-coding-system (coding-system)
1271 "Make CODING-SYSTEM used for communicating with other X clients.
1272 When sending or receiving text via cut_buffer, selection, and clipboard,
1273 the text is encoded or decoded by CODING-SYSTEM."
1274 (interactive "zCoding system for X selection: ")
1275 (check-coding-system coding-system
)
1276 (setq selection-coding-system coding-system
))
1278 ;; Coding system lastly specified by the command
1279 ;; set-next-selection-coding-system.
1280 (defvar last-next-selection-coding-system nil
)
1282 (defun set-next-selection-coding-system (coding-system)
1283 "Make CODING-SYSTEM used for the next communication with other X clients.
1284 This setting is effective for the next communication only."
1286 (list (read-coding-system
1287 (if last-next-selection-coding-system
1288 (format "Coding system for the next X selection (default, %S): "
1289 last-next-selection-coding-system
)
1290 "Coding system for the next X selection: ")
1291 last-next-selection-coding-system
)))
1293 (setq last-next-selection-coding-system coding-system
)
1294 (setq coding-system last-next-selection-coding-system
))
1295 (check-coding-system coding-system
)
1297 (setq next-selection-coding-system coding-system
))
1299 (defun set-coding-priority (arg)
1300 "Set priority of coding categories according to ARG.
1301 ARG is a list of coding categories ordered by priority."
1303 (current-list (copy-sequence coding-category-list
)))
1304 ;; Check the validity of ARG while deleting coding categories in
1305 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
1306 ;; contains all coding categories.
1308 (if (or (null (get (car l
) 'coding-category-index
))
1309 (null (memq (car l
) current-list
)))
1310 (error "Invalid or duplicated element in argument: %s" arg
))
1311 (setq current-list
(delq (car l
) current-list
))
1313 ;; Update `coding-category-list' and return it.
1314 (setq coding-category-list
(append arg current-list
))
1315 (set-coding-priority-internal)))
1319 (defvar non-standard-icccm-encodings-alist
1320 '(("ISO8859-15" . latin-iso8859-15
)
1321 ("ISO8859-14" . latin-iso8859-14
)
1324 "Alist of font charset names defined by XLFD.
1325 The cdr of each element is the corresponding Emacs charset or coding system.")
1327 ;; Functions to support "Non-Standard Character Set Encodings" defined
1328 ;; by the COMPOUND-TEXT spec.
1329 ;; We support that by converting the leading sequence of the
1330 ;; ``extended segment'' to the corresponding ISO-2022 sequences (if
1331 ;; the leading sequence names an Emacs charset), or decode the segment
1332 ;; (if it names a coding system). Encoding does the reverse.
1333 ;; This function also supports "The UTF-8 encoding" described in the
1334 ;; section 7 of the documentation fo COMPOUND-TEXT distributed with
1337 (defun ctext-post-read-conversion (len)
1338 "Decode LEN characters encoded as Compound Text with Extended Segments."
1339 (buffer-disable-undo) ; minimize consing due to insertions and deletions
1340 (narrow-to-region (point) (+ (point) len
))
1342 (let ((pt (point-marker))
1343 (oldpt (point-marker))
1344 (newpt (make-marker))
1345 (modified-p (buffer-modified-p))
1346 (case-fold-search nil
)
1347 ;; We need multibyte conversion of "TO" type because the
1348 ;; buffer may be multibyte, and, in that case, the pattern
1349 ;; must contain eight-bit-control/graphic characters.
1350 (pattern (string-to-multibyte "\\(\e\\)%/[0-4]\\([\200-\377][\200-\377]\\)\\([^\002]+\\)\002\\|\e%G[^\e]+\e%@"))
1351 last-coding-system-used
1352 encoding textlen chset
)
1353 (while (re-search-forward pattern nil
'move
)
1354 (set-marker newpt
(point))
1355 (set-marker pt
(match-beginning 0))
1356 (if (= (preceding-char) ?
@)
1357 ;; We found embedded utf-8 sequence.
1359 (delete-char -
3) ; delete ESC % @ at the tail
1361 (delete-char 3) ; delete ESC % G at the head
1363 (decode-coding-region oldpt pt
'ctext-no-compositions
))
1364 (decode-coding-region pt newpt
'mule-utf-8
)
1366 (set-marker oldpt newpt
))
1367 (setq encoding
(match-string 3))
1368 (setq textlen
(- (+ (* (- (aref (match-string 2) 0) 128) 128)
1369 (- (aref (match-string 2) 1) 128))
1370 (1+ (length encoding
))))
1372 chset
(cdr (assoc-ignore-case encoding
1373 non-standard-icccm-encodings-alist
)))
1375 ;; This charset is not supported--leave this extended
1376 ;; segment unaltered and skip over it.
1377 (goto-char (+ (point) textlen
)))
1379 ;; If it's a charset, replace the leading escape sequence
1380 ;; with a standard ISO-2022 sequence. We will decode all
1381 ;; such segments later, in one go, when we exit the loop
1382 ;; or find an extended segment that names a coding
1383 ;; system, not a charset.
1386 (if (= 0 (charset-iso-graphic-plane chset
))
1388 (if (= 1 (charset-dimension chset
)) "(" "$(")
1390 (if (= 96 (charset-chars chset
))
1392 (if (= 1 (charset-dimension chset
)) ")" "$)")))
1393 (string (charset-iso-final-char chset
)))
1395 (goto-char (+ (point) textlen
)))
1396 ((coding-system-p chset
)
1397 ;; If it's a coding system, we need to decode the segment
1398 ;; right away. But first, decode what we've skipped
1399 ;; across until now.
1401 (decode-coding-region oldpt pt
'ctext-no-compositions
))
1402 (delete-region pt newpt
)
1403 (set-marker newpt
(+ newpt textlen
))
1404 (decode-coding-region pt newpt chset
)
1406 (set-marker oldpt newpt
)))))
1407 ;; Decode what's left.
1408 (when (> (point) oldpt
)
1409 (decode-coding-region oldpt
(point) 'ctext-no-compositions
))
1410 ;; This buffer started as unibyte, because the string we get from
1411 ;; the X selection is a unibyte string. We must now make it
1412 ;; multibyte, so that the decoded text is inserted as multibyte
1414 (set-buffer-multibyte t
)
1415 (set-buffer-modified-p modified-p
)
1416 (- (point-max) (point-min)))))
1418 ;; If you add charsets here, be sure to modify the regexp used by
1419 ;; ctext-pre-write-conversion to look up non-standard charsets.
1420 (defvar non-standard-designations-alist
1421 '(("$(0" .
(big5 "big5-0" 2))
1422 ("$(1" .
(big5 "big5-0" 2))
1423 ;; The following are actually standard; generating extended
1424 ;; segments for them is wrong and screws e.g. Latin-9 users.
1425 ;; 8859-{10,13,16} aren't Emacs charsets anyhow. -- fx
1426 ;; ("-V" . (t "iso8859-10" 1))
1427 ;; ("-Y" . (t "iso8859-13" 1))
1428 ;; ("-_" . (t "iso8859-14" 1))
1429 ;; ("-b" . (t "iso8859-15" 1))
1430 ;; ("-f" . (t "iso8859-16" 1))
1432 "Alist of ctext control sequences that introduce character sets which
1433 are not in the list of approved encodings, and the corresponding
1434 coding system, identifier string, and number of octets per encoded
1437 Each element has the form (CTLSEQ . (ENCODING CHARSET NOCTETS)). CTLSEQ
1438 is the control sequence (sans the leading ESC) that introduces the character
1439 set in the text encoded by compound-text. ENCODING is a coding system
1440 symbol; if it is t, it means that the ctext coding system already encodes
1441 the text correctly, and only the leading control sequence needs to be altered.
1442 If ENCODING is a coding system, we need to re-encode the text with that
1443 coding system. CHARSET is the name of the charset we need to put into
1444 the leading control sequence. NOCTETS is the number of octets (bytes) that
1445 encode each character in this charset. NOCTETS can be 0 (meaning the number
1446 of octets per character is variable), 1, 2, 3, or 4.")
1448 (defun ctext-pre-write-conversion (from to
)
1449 "Encode characters between FROM and TO as Compound Text w/Extended Segments.
1451 If FROM is a string, or if the current buffer is not the one set up for us
1452 by run_pre_post_conversion_on_str, generate a new temp buffer, insert the
1453 text, and convert it in the temporary buffer. Otherwise, convert in-place."
1454 (cond ((and (string= (buffer-name) " *code-converting-work*")
1455 (not (stringp from
)))
1456 ; Minimize consing due to subsequent insertions and deletions.
1457 (buffer-disable-undo)
1458 (narrow-to-region from to
))
1460 (let ((buf (current-buffer)))
1461 (set-buffer (generate-new-buffer " *temp"))
1462 (buffer-disable-undo)
1465 (insert-buffer-substring buf from to
))
1466 (setq from
(point-min) to
(point-max)))))
1467 (encode-coding-region from to
'ctext-no-compositions
)
1468 ;; Replace ISO-2022 charset designations with extended segments, for
1469 ;; those charsets that are not part of the official X registry.
1471 (goto-char (point-min))
1472 (let ((newpt (make-marker))
1473 (case-fold-search nil
)
1474 pt desig encode-info encoding chset noctets textlen
)
1475 (set-buffer-multibyte nil
)
1476 ;; The regexp below finds the leading sequences for big5.
1477 (while (re-search-forward "\e\\(\$([01]\\)" nil
'move
)
1478 (setq desig
(match-string 1)
1480 encode-info
(cdr (assoc desig non-standard-designations-alist
))
1481 encoding
(car encode-info
)
1482 chset
(cadr encode-info
)
1483 noctets
(car (cddr encode-info
)))
1484 (skip-chars-forward "^\e")
1485 (set-marker newpt
(point))
1487 ((eq encoding t
) ; only the leading sequence needs to be changed
1488 (setq textlen
(+ (- newpt pt
) (length chset
) 1))
1489 ;; Generate the ICCCM control sequence for an extended segment.
1490 (replace-match (format "\e%%/%d%c%c%s\x02"
1492 (+ (/ textlen
128) 128)
1493 (+ (% textlen
128) 128)
1496 ((coding-system-p encoding
) ; need to recode the entire segment...
1497 (set-marker pt
(match-beginning 0))
1498 (decode-coding-region pt newpt
'ctext-no-compositions
)
1499 (set-buffer-multibyte t
)
1500 (encode-coding-region pt newpt encoding
)
1501 (set-buffer-multibyte nil
)
1502 (setq textlen
(+ (- newpt pt
) (length chset
) 1))
1504 (insert (format "\e%%/%d%c%c%s\x02"
1506 (+ (/ textlen
128) 128)
1507 (+ (% textlen
128) 128)
1509 (goto-char newpt
))))
1510 (set-buffer-multibyte t
)
1511 ;; Must return nil, as build_annotations_2 expects that.
1516 (defcustom auto-coding-alist
1517 '(("\\.\\(arc\\|zip\\|lzh\\|zoo\\|jar\\|sx[dmicw]\\|tar\\|tgz\\)\\'" . no-conversion
)
1518 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|gpg\\)\\'" . no-conversion
)
1519 ("/#[^/]+#\\'" . emacs-mule
))
1520 "Alist of filename patterns vs corresponding coding systems.
1521 Each element looks like (REGEXP . CODING-SYSTEM).
1522 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1524 The settings in this alist take priority over `coding:' tags
1525 in the file (see the function `set-auto-coding')
1526 and the contents of `file-coding-system-alist'."
1529 :type
'(repeat (cons (regexp :tag
"File name regexp")
1530 (symbol :tag
"Coding system"))))
1532 (defcustom auto-coding-regexp-alist
1533 '(("^BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion
))
1534 "Alist of patterns vs corresponding coding systems.
1535 Each element looks like (REGEXP . CODING-SYSTEM).
1536 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1538 The settings in this alist take priority over `coding:' tags
1539 in the file (see the function `set-auto-coding')
1540 and the contents of `file-coding-system-alist'."
1543 :type
'(repeat (cons (regexp :tag
"Regexp")
1544 (symbol :tag
"Coding system"))))
1546 ;; See the bottom of this file for built-in auto coding functions.
1547 (defcustom auto-coding-functions
'(sgml-xml-auto-coding-function
1548 sgml-html-meta-auto-coding-function
)
1549 "A list of functions which attempt to determine a coding system.
1551 Each function in this list should be written to operate on the current
1552 buffer, but should not modify it in any way. It should take one
1553 argument SIZE, past which it should not search. If a function
1554 succeeds in determining a coding system, it should return that coding
1555 system. Otherwise, it should return nil.
1557 Any `coding:' tags present have a higher priority than the
1558 functions in this list."
1561 :type
'(repeat function
))
1563 (defvar set-auto-coding-for-load nil
1564 "Non-nil means look for `load-coding' property instead of `coding'.
1565 This is used for loading and byte-compiling Emacs Lisp files.")
1567 (defun auto-coding-alist-lookup (filename)
1568 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1569 (let ((alist auto-coding-alist
)
1570 (case-fold-search (memq system-type
'(vax-vms windows-nt ms-dos cygwin
)))
1572 (while (and alist
(not coding-system
))
1573 (if (string-match (car (car alist
)) filename
)
1574 (setq coding-system
(cdr (car alist
)))
1575 (setq alist
(cdr alist
))))
1578 (defun set-auto-coding (filename size
)
1579 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1580 These bytes should include at least the first 1k of the file
1581 and the last 3k of the file, but the middle may be omitted.
1583 The function checks FILENAME against the variable `auto-coding-alist'.
1584 If FILENAME doesn't match any entries in the variable, it checks the
1585 contents of the current buffer following point against
1586 `auto-coding-regexp-alist'. If no match is found, it checks for a
1587 `coding:' tag in the first one or two lines following point. If no
1588 `coding:' tag is found, it checks any local variables list in the last
1589 3K bytes out of the SIZE bytes. Finally, if none of these methods
1590 succeed, it checks to see if any function in `auto-coding-functions'
1593 The return value is the specified coding system, or nil if nothing is
1596 The variable `set-auto-coding-function' (which see) is set to this
1597 function by default."
1598 (or (auto-coding-alist-lookup filename
)
1599 ;; Try using `auto-coding-regexp-alist'.
1601 (let ((alist auto-coding-regexp-alist
)
1603 (while (and alist
(not coding-system
))
1604 (let ((regexp (car (car alist
))))
1605 (when (re-search-forward regexp
(+ (point) size
) t
)
1606 (setq coding-system
(cdr (car alist
)))))
1607 (setq alist
(cdr alist
)))
1609 (let* ((case-fold-search t
)
1610 (head-start (point))
1611 (head-end (+ head-start
(min size
1024)))
1612 (tail-start (+ head-start
(max (- size
3072) 0)))
1613 (tail-end (+ head-start size
))
1614 coding-system head-found tail-found pos
)
1615 ;; Try a short cut by searching for the string "coding:"
1616 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1617 (setq head-found
(or (search-forward "coding:" head-end t
)
1618 (search-forward "unibyte:" head-end t
)))
1619 (if (and head-found
(> head-found tail-start
))
1620 ;; Head and tail are overlapped.
1621 (setq tail-found head-found
)
1622 (goto-char tail-start
)
1623 (setq tail-found
(or (search-forward "coding:" tail-end t
)
1624 (search-forward "unibyte:" tail-end t
))))
1626 ;; At first check the head.
1628 (goto-char head-start
)
1629 (setq head-end
(set-auto-mode-1))
1630 (setq head-start
(point))
1631 (when (and head-end
(< head-found head-end
))
1632 (goto-char head-start
)
1633 (when (and set-auto-coding-for-load
1635 "\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1637 (setq coding-system
'raw-text
))
1638 (when (and (not coding-system
)
1640 "\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1642 (setq coding-system
(intern (match-string 2)))
1643 (or (coding-system-p coding-system
)
1644 (setq coding-system nil
)))))
1646 ;; If no coding: tag in the head, check the tail.
1647 (when (and tail-found
(not coding-system
))
1648 (goto-char tail-start
)
1649 (search-forward "\n\^L" nil t
)
1650 (if (re-search-forward
1651 "^\\(.*\\)[ \t]*Local Variables:[ \t]*\\(.*\\)$" tail-end t
)
1652 ;; The prefix is what comes before "local variables:" in its
1653 ;; line. The suffix is what comes after "local variables:"
1655 (let* ((prefix (regexp-quote (match-string 1)))
1656 (suffix (regexp-quote (match-string 2)))
1660 ;; N.B. without the \n below, the regexp can
1662 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\n]+\\)[ \t]*"
1667 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\n]+\\)[ \t]*"
1670 (concat "^" prefix
"[ \t]*End *:[ \t]*" suffix
"$"))
1672 (re-search-forward re-end tail-end
'move
)
1673 (setq tail-end
(point))
1675 (when (and set-auto-coding-for-load
1676 (re-search-forward re-unibyte tail-end t
))
1677 (setq coding-system
'raw-text
))
1678 (when (and (not coding-system
)
1679 (re-search-forward re-coding tail-end t
))
1680 (setq coding-system
(intern (match-string 1)))
1681 (or (coding-system-p coding-system
)
1682 (setq coding-system nil
))))))
1684 ;; Finally, try all the `auto-coding-functions'.
1685 (let ((funcs auto-coding-functions
)
1686 (coding-system nil
))
1687 (while (and funcs
(not coding-system
))
1688 (setq coding-system
(condition-case e
1690 (goto-char (point-min))
1691 (funcall (pop funcs
) size
))
1695 (setq set-auto-coding-function
'set-auto-coding
)
1697 (defun after-insert-file-set-buffer-file-coding-system (inserted)
1698 "Set `buffer-file-coding-system' of current buffer after text is inserted."
1699 (if last-coding-system-used
1700 (let ((coding-system
1701 (find-new-buffer-file-coding-system last-coding-system-used
))
1702 (modified-p (buffer-modified-p)))
1704 (set-buffer-file-coding-system coding-system t
)
1705 (if (and enable-multibyte-characters
1706 (or (eq coding-system
'no-conversion
)
1707 (eq (coding-system-type coding-system
) 5))
1708 ;; If buffer was unmodified and the size is the
1709 ;; same as INSERTED, we must be visiting it.
1711 (= (buffer-size) inserted
))
1712 ;; For coding systems no-conversion and raw-text...,
1713 ;; edit the buffer as unibyte.
1714 (let ((pos-byte (position-bytes (+ (point) inserted
))))
1715 (set-buffer-multibyte nil
)
1716 (setq inserted
(- pos-byte
(position-bytes (point))))))
1717 (set-buffer-modified-p modified-p
))))
1720 (setq after-insert-file-adjust-coding-function
1721 'after-insert-file-set-buffer-file-coding-system
)
1723 ;; The coding-spec and eol-type of coding-system returned is decided
1724 ;; independently in the following order.
1725 ;; 1. That of buffer-file-coding-system locally bound.
1726 ;; 2. That of CODING.
1728 (defun find-new-buffer-file-coding-system (coding)
1729 "Return a coding system for a buffer when a file of CODING is inserted.
1730 The local variable `buffer-file-coding-system' of the current buffer
1731 is set to the returned value.
1732 Return nil if there's no need to set `buffer-file-coding-system'."
1733 (let (local-coding local-eol
1734 found-coding found-eol
1737 ;; Nothing found about coding.
1740 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1741 ;; and LOCAL-CODING.
1742 (setq local-eol
(coding-system-eol-type buffer-file-coding-system
))
1743 (if (null (numberp local-eol
))
1744 ;; But eol-type is not yet set.
1745 (setq local-eol nil
))
1746 (if (and buffer-file-coding-system
1747 (not (eq (coding-system-type buffer-file-coding-system
) t
)))
1748 ;; This is not `undecided'.
1749 (setq local-coding
(coding-system-base buffer-file-coding-system
)))
1751 (if (and (local-variable-p 'buffer-file-coding-system
)
1752 local-eol local-coding
)
1753 ;; The current buffer has already set full coding-system, we
1754 ;; had better not change it.
1757 (setq found-eol
(coding-system-eol-type coding
))
1758 (if (null (numberp found-eol
))
1759 ;; But eol-type is not found.
1760 ;; If EOL conversions are inhibited, force unix eol-type.
1761 (setq found-eol
(if inhibit-eol-conversion
0)))
1762 (if (eq (coding-system-type coding
) t
)
1763 (setq found-coding
'undecided
)
1764 (setq found-coding
(coding-system-base coding
)))
1766 (if (and (not found-eol
) (eq found-coding
'undecided
))
1767 ;; No valid coding information found.
1770 ;; Some coding information (eol or text) found.
1772 ;; The local setting takes precedence over the found one.
1773 (setq new-coding
(if (local-variable-p 'buffer-file-coding-system
)
1774 (or local-coding found-coding
)
1775 (or found-coding local-coding
)))
1776 (setq new-eol
(if (local-variable-p 'buffer-file-coding-system
)
1777 (or local-eol found-eol
)
1778 (or found-eol local-eol
)))
1780 (let ((eol-type (coding-system-eol-type new-coding
)))
1781 (if (and (numberp new-eol
) (vectorp eol-type
))
1782 (aref eol-type new-eol
)
1785 (defun modify-coding-system-alist (target-type regexp coding-system
)
1786 "Modify one of look up tables for finding a coding system on I/O operation.
1787 There are three of such tables, `file-coding-system-alist',
1788 `process-coding-system-alist', and `network-coding-system-alist'.
1790 TARGET-TYPE specifies which of them to modify.
1791 If it is `file', it affects `file-coding-system-alist' (which see).
1792 If it is `process', it affects `process-coding-system-alist' (which see).
1793 If it is `network', it affects `network-coding-system-alist' (which see).
1795 REGEXP is a regular expression matching a target of I/O operation.
1796 The target is a file name if TARGET-TYPE is `file', a program name if
1797 TARGET-TYPE is `process', or a network service name or a port number
1798 to connect to if TARGET-TYPE is `network'.
1800 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1801 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1802 for decoding and encoding respectively,
1803 or a function symbol which, when called, returns such a cons cell."
1804 (or (memq target-type
'(file process network
))
1805 (error "Invalid target type: %s" target-type
))
1806 (or (stringp regexp
)
1807 (and (eq target-type
'network
) (integerp regexp
))
1808 (error "Invalid regular expression: %s" regexp
))
1809 (if (symbolp coding-system
)
1810 (if (not (fboundp coding-system
))
1812 (check-coding-system coding-system
)
1813 (setq coding-system
(cons coding-system coding-system
))))
1814 (check-coding-system (car coding-system
))
1815 (check-coding-system (cdr coding-system
)))
1816 (cond ((eq target-type
'file
)
1817 (let ((slot (assoc regexp file-coding-system-alist
)))
1819 (setcdr slot coding-system
)
1820 (setq file-coding-system-alist
1821 (cons (cons regexp coding-system
)
1822 file-coding-system-alist
)))))
1823 ((eq target-type
'process
)
1824 (let ((slot (assoc regexp process-coding-system-alist
)))
1826 (setcdr slot coding-system
)
1827 (setq process-coding-system-alist
1828 (cons (cons regexp coding-system
)
1829 process-coding-system-alist
)))))
1831 (let ((slot (assoc regexp network-coding-system-alist
)))
1833 (setcdr slot coding-system
)
1834 (setq network-coding-system-alist
1835 (cons (cons regexp coding-system
)
1836 network-coding-system-alist
)))))))
1838 (defun decode-coding-inserted-region (from to filename
1839 &optional visit beg end replace
)
1840 "Decode the region between FROM and TO as if it is read from file FILENAME.
1841 Optional arguments VISIT, BEG, END, and REPLACE are the same as those
1842 of the function `insert-file-contents'."
1845 (narrow-to-region from to
)
1846 (goto-char (point-min))
1847 (let ((coding coding-system-for-read
))
1849 (setq coding
(funcall set-auto-coding-function
1850 filename
(- (point-max) (point-min)))))
1852 (setq coding
(find-operation-coding-system
1853 'insert-file-contents
1854 filename visit beg end replace
)))
1855 (if (coding-system-p coding
)
1856 (or enable-multibyte-characters
1858 (coding-system-change-text-conversion coding
'raw-text
)))
1861 (decode-coding-region (point-min) (point-max) coding
))
1862 (setq last-coding-system-used coding
)))))
1864 (defun make-translation-table (&rest args
)
1865 "Make a translation table from arguments.
1866 A translation table is a char table intended for character
1867 translation in CCL programs.
1869 Each argument is a list of elements of the form (FROM . TO), where FROM
1870 is a character to be translated to TO.
1872 FROM can be a generic character (see `make-char'). In this case, TO is
1873 a generic character containing the same number of characters, or an
1874 ordinary character. If FROM and TO are both generic characters, all
1875 characters belonging to FROM are translated to characters belonging to TO
1876 without changing their position code(s).
1878 The arguments and forms in each argument are processed in the given
1879 order, and if a previous form already translates TO to some other
1880 character, say TO-ALT, FROM is also translated to TO-ALT."
1881 (let ((table (make-char-table 'translation-table
))
1884 (let ((elts (car args
)))
1886 (let* ((from (car (car elts
)))
1887 (from-i 0) ; degree of freedom of FROM
1888 (from-rev (nreverse (split-char from
)))
1889 (to (cdr (car elts
)))
1890 (to-i 0) ; degree of freedom of TO
1891 (to-rev (nreverse (split-char to
))))
1892 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
1893 (while (eq (car from-rev
) 0)
1894 (setq from-i
(1+ from-i
) from-rev
(cdr from-rev
)))
1895 (while (eq (car to-rev
) 0)
1896 (setq to-i
(1+ to-i
) to-rev
(cdr to-rev
)))
1897 (if (and (/= from-i to-i
) (/= to-i
0))
1898 (error "Invalid character pair (%d . %d)" from to
))
1899 ;; If we have already translated TO to TO-ALT, FROM should
1900 ;; also be translated to TO-ALT. But, this is only if TO
1901 ;; is a generic character or TO-ALT is not a generic
1903 (let ((to-alt (aref table to
)))
1905 (or (> to-i
0) (not (generic-char-p to-alt
))))
1908 (set-char-table-default table from to
)
1909 (aset table from to
))
1910 ;; If we have already translated some chars to FROM, they
1911 ;; should also be translated to TO.
1912 (let ((l (assq from revlist
)))
1919 (setq l
(cdr l
)) ))))
1920 ;; Now update REVLIST.
1921 (let ((l (assq to revlist
)))
1923 (setcdr l
(cons from
(cdr l
)))
1924 (setq revlist
(cons (list to from
) revlist
)))))
1925 (setq elts
(cdr elts
))))
1926 (setq args
(cdr args
)))
1927 ;; Return TABLE just created.
1930 (defun make-translation-table-from-vector (vec)
1931 "Make translation table from decoding vector VEC.
1932 VEC is an array of 256 elements to map unibyte codes to multibyte
1933 characters. Elements may be nil for undefined code points.
1934 See also the variable `nonascii-translation-table'."
1935 (let ((table (make-char-table 'translation-table
))
1936 (rev-table (make-char-table 'translation-table
))
1939 (setq ch
(aref vec i
))
1943 (aset rev-table ch i
))))
1944 (set-char-table-extra-slot table
0 rev-table
)
1947 (defun define-translation-table (symbol &rest args
)
1948 "Define SYMBOL as the name of translation table made by ARGS.
1949 This sets up information so that the table can be used for
1950 translations in a CCL program.
1952 If the first element of ARGS is a char-table whose purpose is
1953 `translation-table', just define SYMBOL to name it. (Note that this
1954 function does not bind SYMBOL.)
1956 Any other ARGS should be suitable as arguments of the function
1957 `make-translation-table' (which see).
1959 This function sets properties `translation-table' and
1960 `translation-table-id' of SYMBOL to the created table itself and the
1961 identification number of the table respectively. It also registers
1962 the table in `translation-table-vector'."
1963 (let ((table (if (and (char-table-p (car args
))
1964 (eq (char-table-subtype (car args
))
1965 'translation-table
))
1967 (apply 'make-translation-table args
)))
1968 (len (length translation-table-vector
))
1971 (put symbol
'translation-table table
)
1974 (setq translation-table-vector
1975 (vconcat translation-table-vector
(make-vector len nil
))))
1976 (let ((slot (aref translation-table-vector id
)))
1978 (eq (car slot
) symbol
))
1980 (aset translation-table-vector id
(cons symbol table
))
1982 (setq id
(1+ id
)))))
1983 (put symbol
'translation-table-id id
)
1986 (put 'with-category-table
'lisp-indent-function
1)
1988 (defmacro with-category-table
(table &rest body
)
1989 "Evaluate BODY with category table of current buffer set to TABLE.
1990 The category table of the current buffer is saved, BODY is evaluated,
1991 then the saved table is restored, even in case of an abnormal exit.
1992 Value is what BODY returns."
1993 (let ((old-table (make-symbol "old-table"))
1994 (old-buffer (make-symbol "old-buffer")))
1995 `(let ((,old-table
(category-table))
1996 (,old-buffer
(current-buffer)))
1999 (set-category-table ,table
)
2001 (save-current-buffer
2002 (set-buffer ,old-buffer
)
2003 (set-category-table ,old-table
))))))
2005 (defun define-translation-hash-table (symbol table
)
2006 "Define SYMBOL as the name of the hash translation TABLE for use in CCL.
2008 Analogous to `define-translation-table', but updates
2009 `translation-hash-table-vector' and the table is for use in the CCL
2010 `lookup-integer' and `lookup-character' functions."
2011 (unless (and (symbolp symbol
)
2012 (hash-table-p table
))
2013 (error "Bad args to define-translation-hash-table"))
2014 (let ((len (length translation-hash-table-vector
))
2017 (put symbol
'translation-hash-table table
)
2020 (setq translation-hash-table-vector
2021 (vconcat translation-hash-table-vector
[nil])))
2022 (let ((slot (aref translation-hash-table-vector id)))
2024 (eq (car slot) symbol))
2026 (aset translation-hash-table-vector id (cons symbol table))
2028 (setq id (1+ id)))))
2029 (put symbol 'translation-hash-table-id id)
2032 ;;; Initialize some variables.
2034 (put 'use-default-ascent 'char-table-extra-slots 0)
2035 (setq use-default-ascent (make-char-table 'use-default-ascent))
2036 (put 'ignore-relative-composition 'char-table-extra-slots 0)
2037 (setq ignore-relative-composition
2038 (make-char-table 'ignore-relative-composition))
2041 ;;; Built-in auto-coding-functions:
2043 (defun sgml-xml-auto-coding-function (size)
2044 "Determine whether the buffer is XML, and if so, its encoding.
2045 This function is intended to be added to `auto-coding-functions'."
2046 (when (re-search-forward "\\`[[:space:]\n]*<\\?xml" nil t)
2047 (let ((end (save-excursion
2049 (re-search-forward "\"\\s-*\\?>" size t))))
2051 (if (re-search-forward "encoding=\"\\(.+?\\)\"" end t)
2052 (let* ((match (match-string 1))
2053 (sym (intern (downcase match))))
2054 (if (coding-system-p sym)
2056 (message "Warning: unknown coding system \"%s\"" match)
2060 (defun sgml-html-meta-auto-coding-function (size)
2061 "If the buffer has an HTML meta tag, use it to determine encoding.
2062 This function is intended to be added to `auto-coding-functions'."
2063 (setq size (min size
2064 ;; Only search forward 10 lines
2068 (when (and (search-forward "<html>" size t)
2069 (re-search-forward "<meta\\s-+http-equiv=\"content-type\"\\s-+content=\"text/\\sw+;\\s-*charset=\\(.+?\\)\"" size t))
2070 (let* ((match (match-string 1))
2071 (sym (intern (downcase match))))
2072 (if (coding-system-p sym)
2074 (message "Warning: unknown coding system \"%s\"" match)
2080 ;;; mule.el ends here