1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Copyright (C) 2001 Free Software Foundation, Inc.
5 ;; Licensed to the Free Software Foundation.
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 (setq preloaded-file-list (cons 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 error."
273 (make-char-internal (charset-id charset) code1 code2))
275 (put 'make-char 'byte-compile
278 (let ((charset (nth 1 form)))
279 (if (charset-quoted-standard-p charset)
280 (byte-compile-normal-call
281 (cons 'make-char-internal
282 (cons (charset-id (nth 1 charset)) (nthcdr 2 form))))
283 (byte-compile-normal-call
284 (cons 'make-char-internal
285 (cons (list 'charset-id charset) (nthcdr 2 form)))))))))
287 (defun charset-list ()
288 "Return list of charsets ever defined.
290 This function is provided for backward compatibility.
291 Now we have the variable `charset-list'."
294 (defsubst generic-char-p (char)
295 "Return t if and only if CHAR is a generic character.
296 See also the documentation of `make-char'."
298 (let ((l (split-char char)))
299 (and (or (= (nth 1 l) 0) (eq (nth 2 l) 0))
300 (not (eq (car l) 'composition))))))
302 (defun decode-char (ccs code-point &optional restriction)
303 "Return character specified by coded character set CCS and CODE-POINT in it.
304 Return nil if such a character is not supported.
305 Currently the only supported coded character set is `ucs' (ISO/IEC
306 10646: Universal Multi-Octet Coded Character Set).
308 Optional argument RESTRICTION specifies a way to map the pair of CCS
309 and CODE-POINT to a character. Currently not supported and just ignored."
311 (cond ((< code-point 160)
314 (make-char 'latin-iso8859-1 code-point))
315 ((< code-point #x2500)
316 (setq code-point (- code-point #x0100))
317 (make-char 'mule-unicode-0100-24ff
318 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
319 ((< code-point #x3400)
320 (setq code-point (- code-point #x2500))
321 (make-char 'mule-unicode-2500-33ff
322 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
323 ((and (>= code-point #xe000) (< code-point #x10000))
324 (setq code-point (- code-point #xe000))
325 (make-char 'mule-unicode-e000-ffff
326 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
329 (defun encode-char (char ccs &optional restriction)
330 "Return code-point in coded character set CCS that corresponds to CHAR.
331 Return nil if CHAR is not included in CCS.
332 Currently the only supported coded character set is `ucs' (ISO/IEC
333 10646: Universal Multi-Octet Coded Character Set).
335 CHAR should be in one of these charsets:
336 ascii, latin-iso8859-1, mule-unicode-0100-24ff, mule-unicode-2500-33ff,
337 mule-unicode-e000-ffff, eight-bit-control
338 Otherwise, return nil.
340 Optional argument RESTRICTION specifies a way to map CHAR to a
341 code-point in CCS. Currently not supported and just ignored."
342 (let* ((split (split-char char))
343 (charset (car split)))
345 (cond ((eq charset 'ascii)
347 ((eq charset 'latin-iso8859-1)
348 (+ (nth 1 split) 128))
349 ((eq charset 'mule-unicode-0100-24ff)
350 (+ #x0100 (+ (* (- (nth 1 split) 32) 96)
351 (- (nth 2 split) 32))))
352 ((eq charset 'mule-unicode-2500-33ff)
353 (+ #x2500 (+ (* (- (nth 1 split) 32) 96)
354 (- (nth 2 split) 32))))
355 ((eq charset 'mule-unicode-e000-ffff)
356 (+ #xe000 (+ (* (- (nth 1 split) 32) 96)
357 (- (nth 2 split) 32))))
358 ((eq charset 'eight-bit-control)
362 ;; Coding system stuff
364 ;; Coding system is a symbol that has the property `coding-system'.
366 ;; The value of the property `coding-system' is a vector of the
368 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
369 ;; We call this vector as coding-spec. See comments in src/coding.c
372 (defconst coding-spec-type-idx 0)
373 (defconst coding-spec-mnemonic-idx 1)
374 (defconst coding-spec-doc-string-idx 2)
375 (defconst coding-spec-plist-idx 3)
376 (defconst coding-spec-flags-idx 4)
378 ;; PLIST is a property list of a coding system. To share PLIST among
379 ;; alias coding systems, a coding system has PLIST in coding-spec
380 ;; instead of having it in normal property list of Lisp symbol.
381 ;; Here's a list of coding system properties currently being used.
385 ;; The value is a coding category the coding system belongs to. The
386 ;; function `make-coding-system' sets this value automatically
387 ;; unless its argument PROPERTIES specifies this property.
389 ;; o alias-coding-systems
391 ;; The value is a list of coding systems of the same alias group. The
392 ;; first element is the coding system made at first, which we call as
393 ;; `base coding system'. The function `make-coding-system' sets this
394 ;; value automatically and `define-coding-system-alias' updates it.
396 ;; See the documentation of make-coding-system for the meanings of the
397 ;; following properties.
399 ;; o post-read-conversion
400 ;; o pre-write-conversion
401 ;; o translation-table-for-decode
402 ;; o translation-table-for-encode
406 ;; o valid-codes (meaningful only for a coding system based on CCL)
409 (defsubst coding-system-spec (coding-system)
410 "Return coding-spec of CODING-SYSTEM."
411 (get (check-coding-system coding-system) 'coding-system))
413 (defun coding-system-type (coding-system)
414 "Return the coding type of CODING-SYSTEM.
415 A coding type is an integer value indicating the encoding method
416 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
417 (aref (coding-system-spec coding-system) coding-spec-type-idx))
419 (defun coding-system-mnemonic (coding-system)
420 "Return the mnemonic character of CODING-SYSTEM.
421 The mnemonic character of a coding system is used in mode line
422 to indicate the coding system. If the arg is nil, return ?-."
423 (let ((spec (coding-system-spec coding-system)))
424 (if spec (aref spec coding-spec-mnemonic-idx) ?-)))
426 (defun coding-system-doc-string (coding-system)
427 "Return the documentation string for CODING-SYSTEM."
428 (aref (coding-system-spec coding-system) coding-spec-doc-string-idx))
430 (defun coding-system-plist (coding-system)
431 "Return the property list of CODING-SYSTEM."
432 (aref (coding-system-spec coding-system) coding-spec-plist-idx))
434 (defun coding-system-flags (coding-system)
435 "Return `flags' of CODING-SYSTEM.
436 A `flags' of a coding system is a vector of length 32 indicating detailed
437 information of a coding system. See the function `make-coding-system'
439 (aref (coding-system-spec coding-system) coding-spec-flags-idx))
441 (defun coding-system-get (coding-system prop)
442 "Extract a value from CODING-SYSTEM's property list for property PROP."
443 (plist-get (coding-system-plist coding-system) prop))
445 (defun coding-system-put (coding-system prop val)
446 "Change value in CODING-SYSTEM's property list PROP to VAL."
447 (let ((plist (coding-system-plist coding-system)))
449 (plist-put plist prop val)
450 (aset (coding-system-spec coding-system) coding-spec-plist-idx
453 (defun coding-system-category (coding-system)
454 "Return the coding category of CODING-SYSTEM.
455 See also `coding-category-list'."
456 (coding-system-get coding-system 'coding-category))
458 (defun coding-system-base (coding-system)
459 "Return the base coding system of CODING-SYSTEM.
460 A base coding system is what made by `make-coding-system'.
461 Any alias nor subsidiary coding systems are not base coding system."
462 (car (coding-system-get coding-system 'alias-coding-systems)))
464 (defalias 'coding-system-parent 'coding-system-base)
465 (make-obsolete 'coding-system-parent 'coding-system-base "20.3")
467 ;; Coding system also has a property `eol-type'.
469 ;; This property indicates how the coding system handles end-of-line
470 ;; format. The value is integer 0, 1, 2, or a vector of three coding
471 ;; systems. Each integer value 0, 1, and 2 indicates the format of
472 ;; end-of-line LF, CRLF, and CR respectively. A vector value
473 ;; indicates that the format of end-of-line should be detected
474 ;; automatically. Nth element of the vector is the subsidiary coding
475 ;; system whose `eol-type' property is N.
477 (defun coding-system-eol-type (coding-system)
478 "Return eol-type of CODING-SYSTEM.
479 An eol-type is integer 0, 1, 2, or a vector of coding systems.
481 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
482 CRLF, and CR respectively.
484 A vector value indicates that a format of end-of-line should be
485 detected automatically. Nth element of the vector is the subsidiary
486 coding system whose eol-type is N."
487 (get coding-system 'eol-type))
489 (defun coding-system-lessp (x y)
490 (cond ((eq x 'no-conversion) t)
491 ((eq y 'no-conversion) nil)
492 ((eq x 'emacs-mule) t)
493 ((eq y 'emacs-mule) nil)
494 ((eq x 'undecided) t)
495 ((eq y 'undecided) nil)
496 (t (let ((c1 (coding-system-mnemonic x))
497 (c2 (coding-system-mnemonic y)))
498 (or (< (downcase c1) (downcase c2))
499 (and (not (> (downcase c1) (downcase c2)))
502 (defun add-to-coding-system-list (coding-system)
503 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
504 (if (or (null coding-system-list)
505 (coding-system-lessp coding-system (car coding-system-list)))
506 (setq coding-system-list (cons coding-system coding-system-list))
507 (let ((len (length coding-system-list))
508 mid (tem coding-system-list))
510 (setq mid (nthcdr (/ len 2) tem))
511 (if (coding-system-lessp (car mid) coding-system)
513 len (- len (/ len 2)))
514 (setq len (/ len 2))))
515 (setcdr tem (cons coding-system (cdr tem))))))
517 (defun coding-system-list (&optional base-only)
518 "Return a list of all existing non-subsidiary coding systems.
519 If optional arg BASE-ONLY is non-nil, only base coding systems are listed.
520 The value doesn't include subsidiary coding systems which are what
521 made from bases and aliases automatically for various end-of-line
522 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
523 (let* ((codings (copy-sequence coding-system-list))
524 (tail (cons nil codings)))
525 ;; Remove subsidiary coding systems (eol variants) and alias
526 ;; coding systems (if necessary).
528 (let* ((coding (car (cdr tail)))
529 (aliases (coding-system-get coding 'alias-coding-systems)))
531 ;; CODING is an eol variant if not in ALIASES.
532 (not (memq coding aliases))
533 ;; CODING is an alias if it is not car of ALIASES.
534 (and base-only (not (eq coding (car aliases)))))
535 (setcdr tail (cdr (cdr tail)))
536 (setq tail (cdr tail)))))
539 (defun register-char-codings (coding-system safe-chars)
540 (let ((general (char-table-extra-slot char-coding-system-table 0)))
541 (if (eq safe-chars t)
542 (or (memq coding-system general)
543 (set-char-table-extra-slot char-coding-system-table 0
544 (cons coding-system general)))
548 (if (and (>= key 128) val)
549 (let ((codings (aref char-coding-system-table key)))
550 (or (memq coding-system codings)
551 (aset char-coding-system-table key
552 (cons coding-system codings)))))))
556 (defun make-subsidiary-coding-system (coding-system)
557 "Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM."
558 (let ((coding-spec (coding-system-spec coding-system))
559 (subsidiaries (vector (intern (format "%s-unix" coding-system))
560 (intern (format "%s-dos" coding-system))
561 (intern (format "%s-mac" coding-system))))
565 (put (aref subsidiaries i) 'coding-system coding-spec)
566 (put (aref subsidiaries i) 'eol-type i)
567 (add-to-coding-system-list (aref subsidiaries i))
568 (setq coding-system-alist
569 (cons (list (symbol-name (aref subsidiaries i)))
570 coding-system-alist))
574 (defun transform-make-coding-system-args (name type &optional doc-string props)
575 "For internal use only.
576 Transform XEmacs style args for `make-coding-system' to Emacs style.
577 Value is a list of transformed arguments."
578 (let ((mnemonic (string-to-char (or (plist-get props 'mnemonic) "?")))
579 (eol-type (plist-get props 'eol-type))
582 ((eq eol-type 'lf) (setq eol-type 'unix))
583 ((eq eol-type 'crlf) (setq eol-type 'dos))
584 ((eq eol-type 'cr) (setq eol-type 'mac)))
585 (if (setq tmp (plist-get props 'post-read-conversion))
586 (setq properties (plist-put properties 'post-read-conversion tmp)))
587 (if (setq tmp (plist-get props 'pre-write-conversion))
588 (setq properties (plist-put properties 'pre-write-conversion tmp)))
590 ((eq type 'shift-jis)
591 `(,name 1 ,mnemonic ,doc-string () ,properties ,eol-type))
592 ((eq type 'iso2022) ; This is not perfect.
593 (if (plist-get props 'escape-quoted)
594 (error "escape-quoted is not supported: %S"
595 `(,name ,type ,doc-string ,props)))
596 (let ((g0 (plist-get props 'charset-g0))
597 (g1 (plist-get props 'charset-g1))
598 (g2 (plist-get props 'charset-g2))
599 (g3 (plist-get props 'charset-g3))
602 (eq (cadr (assoc 'latin-jisx0201
603 (plist-get props 'input-charset-conversion)))
605 (eq (cadr (assoc 'ascii
606 (plist-get props 'output-charset-conversion)))
610 (eq (cadr (assoc 'japanese-jisx0208-1978
611 (plist-get props 'input-charset-conversion)))
613 (eq (cadr (assoc 'japanese-jisx0208
614 (plist-get props 'output-charset-conversion)))
615 'japanese-jisx0208-1978))))
617 (if (plist-get props 'force-g0-on-output)
621 (if (plist-get props 'force-g1-on-output)
625 (if (plist-get props 'force-g2-on-output)
629 (if (plist-get props 'force-g3-on-output)
632 `(,name 2 ,mnemonic ,doc-string
634 ,(plist-get props 'short)
635 ,(not (plist-get props 'no-ascii-eol))
636 ,(not (plist-get props 'no-ascii-cntl))
637 ,(plist-get props 'seven)
639 ,(not (plist-get props 'lock-shift))
642 ,(plist-get props 'no-iso6429)
644 ,properties ,eol-type)))
646 `(,name 3 ,mnemonic ,doc-string () ,properties ,eol-type))
648 `(,name 4 ,mnemonic ,doc-string
649 (,(plist-get props 'decode) . ,(plist-get props 'encode))
650 ,properties ,eol-type))
652 (error "unsupported XEmacs style make-coding-style arguments: %S"
653 `(,name ,type ,doc-string ,props))))))
655 (defun make-coding-system (coding-system type mnemonic doc-string
660 "Define a new coding system CODING-SYSTEM (symbol).
661 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
662 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
663 in the following format:
664 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
666 TYPE is an integer value indicating the type of the coding system as follows:
667 0: Emacs internal format,
668 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PC,
669 2: ISO-2022 including many variants,
670 3: Big5 used mainly on Chinese PC,
671 4: private, CCL programs provide encoding/decoding algorithm,
672 5: Raw-text, which means that text contains random 8-bit codes.
674 MNEMONIC is a character to be displayed on mode line for the coding system.
676 DOC-STRING is a documentation string for the coding system.
678 FLAGS specifies more detailed information of the coding system as follows:
680 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
681 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
682 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
683 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
684 SAFE, ACCEPT-LATIN-EXTRA-CODE.
685 CHARSETn are character sets initially designated to Gn graphic registers.
686 If CHARSETn is nil, Gn is never used.
687 If CHARSETn is t, Gn can be used but nothing designated initially.
688 If CHARSETn is a list of character sets, those character sets are
689 designated to Gn on output, but nothing designated to Gn initially.
690 But, character set `ascii' can be designated only to G0.
691 SHORT-FORM non-nil means use short designation sequence on output.
692 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
693 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
695 SEVEN non-nil means use 7-bit code only on output.
696 LOCKING-SHIFT non-nil means use locking-shift.
697 SINGLE-SHIFT non-nil means use single-shift.
698 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
699 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
700 NO-ISO6429 non-nil means not use ISO6429's direction specification.
701 INIT-BOL non-nil means any designation state is assumed to be reset
702 to initial at each beginning of line on output.
703 DESIGNATION-BOL non-nil means designation sequences should be placed
704 at beginning of line on output.
705 SAFE non-nil means convert unsafe characters to `?' on output.
706 Characters not specified in the property `safe-charsets' nor
707 `safe-chars' are unsafe.
708 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
709 a code specified in `latin-extra-code-table' (which see) as a valid
710 code of the coding system.
712 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
713 decoding and encoding. CCL programs should be specified by their
716 PROPERTIES is an alist of properties vs the corresponding values. The
717 following properties are recognized:
719 o post-read-conversion
721 The value is a function to call after some text is inserted and
722 decoded by the coding system itself and before any functions in
723 `after-insert-functions' are called. The argument of this
724 function is the same as for a function in
725 `after-insert-file-functions', i.e. LENGTH of the text inserted,
726 with point at the head of the text to be decoded.
728 o pre-write-conversion
730 The value is a function to call after all functions in
731 `write-region-annotate-functions' and `buffer-file-format' are
732 called, and before the text is encoded by the coding system itself.
733 The arguments to this function are the same as those of a function
734 in `write-region-annotate-functions', i.e. FROM and TO, specifying
737 o translation-table-for-decode
739 The value is a translation table to be applied on decoding. See
740 the function `make-translation-table' for the format of translation
741 table. This is not applicable to type 4 (CCL-based) coding systems.
743 o translation-table-for-encode
745 The value is a translation table to be applied on encoding. This is
746 not applicable to type 4 (CCL-based) coding systems.
750 The value is a char table. If a character has non-nil value in it,
751 the character is safely supported by the coding system. This
752 overrides the specification of safe-charsets.
756 The value is a list of charsets safely supported by the coding
757 system. The value t means that all charsets Emacs handles are
758 supported. Even if some charset is not in this list, it doesn't
759 mean that the charset can't be encoded in the coding system;
760 it just means that some other receiver of text encoded
761 in the coding system won't be able to handle that charset.
765 The value is a symbol of which name is `MIME-charset' parameter of
768 o valid-codes (meaningful only for a coding system based on CCL)
770 The value is a list to indicate valid byte ranges of the encoded
771 file. Each element of the list is an integer or a cons of integer.
772 In the former case, the integer value is a valid byte code. In the
773 latter case, the integers specifies the range of valid byte codes.
775 These properties are set in PLIST, a property list. This function
776 also sets properties `coding-category' and `alias-coding-systems'
779 EOL-TYPE specifies the EOL type of the coding-system in one of the
782 o symbol (unix, dos, or mac)
784 The symbol `unix' means Unix-like EOL (LF), `dos' means
785 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
787 o number (0, 1, or 2)
789 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
792 o vector of coding-systems of length 3
794 The EOL type is detected automatically for the coding system.
795 And, according to the detected EOL type, one of the coding
796 systems in the vector is selected. Elements of the vector
797 corresponds to Unix-like EOL, DOS-like EOL, and Mac-like EOL
800 Kludgy features for backward compatibility:
802 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
803 treated as a compiled CCL code.
805 2. If PROPERTIES is just a list of character sets, the list is set as
806 a value of `safe-charsets' in PLIST."
808 ;; For compatiblity with XEmacs, we check the type of TYPE. If it
809 ;; is a symbol, perhaps, this function is called with XEmacs-style
810 ;; arguments. Here, try to transform that kind of arguments to
813 (let ((args (transform-make-coding-system-args coding-system type
814 mnemonic doc-string)))
815 (setq coding-system (car args)
817 mnemonic (nth 2 args)
818 doc-string (nth 3 args)
820 properties (nth 5 args)
821 eol-type (nth 6 args))))
823 ;; Set a value of `coding-system' property.
824 (let ((coding-spec (make-vector 5 nil))
825 (no-initial-designation t)
826 (no-alternative-designation t)
827 (accept-latin-extra-code nil)
829 (if (or (not (integerp type)) (< type 0) (> type 5))
830 (error "TYPE argument must be 0..5"))
831 (if (or (not (integerp mnemonic)) (<= mnemonic ? ) (> mnemonic 127))
832 (error "MNEMONIC argument must be an ASCII printable character"))
833 (aset coding-spec coding-spec-type-idx type)
834 (aset coding-spec coding-spec-mnemonic-idx mnemonic)
835 (aset coding-spec coding-spec-doc-string-idx
836 (purecopy (if (stringp doc-string) doc-string "")))
838 (setq coding-category 'coding-category-emacs-mule))
840 (setq coding-category 'coding-category-sjis))
841 ((= type 2) ; ISO2022
843 (vec (make-vector 32 nil))
847 (let ((charset (car fl)))
848 (if (and no-initial-designation
850 (or (charsetp charset)
852 (charsetp (car charset)))))
853 (setq no-initial-designation nil))
854 (if (charsetp charset)
855 (if (= i 1) (setq g1-designation charset))
860 (setq elt (car tail))
862 (setq no-alternative-designation nil)
863 (if (and elt (not (charsetp elt)))
864 (error "Invalid charset: %s" elt)))
865 (setq tail (cdr tail)))
866 (setq g1-designation (car charset)))
869 (setq no-alternative-designation nil)
870 (error "Invalid charset: %s" charset)))))
871 (aset vec i charset))
872 (setq fl (cdr fl) i (1+ i)))
873 (while (and (< i 32) fl)
874 (aset vec i (car fl))
875 (if (and (= i 16) ; ACCEPT-LATIN-EXTRA-CODE
877 (setq accept-latin-extra-code t))
878 (setq fl (cdr fl) i (1+ i)))
879 (aset coding-spec 4 vec)
880 (setq coding-category
881 (if (aref vec 8) ; Use locking-shift.
882 (or (and (aref vec 7) 'coding-category-iso-7-else)
883 'coding-category-iso-8-else)
884 (if (aref vec 7) ; 7-bit only.
885 (if (aref vec 9) ; Use single-shift.
886 'coding-category-iso-7-else
887 (if no-alternative-designation
888 'coding-category-iso-7-tight
889 'coding-category-iso-7))
890 (if (or no-initial-designation
891 (not no-alternative-designation))
892 'coding-category-iso-8-else
893 (if (and (charsetp g1-designation)
894 (= (charset-dimension g1-designation) 2))
895 'coding-category-iso-8-2
896 'coding-category-iso-8-1)))))))
898 (setq coding-category 'coding-category-big5))
899 ((= type 4) ; private
900 (setq coding-category 'coding-category-ccl)
901 (if (not (consp flags))
902 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
903 (let ((decoder (check-ccl-program
905 (intern (format "%s-decoder" coding-system))))
906 (encoder (check-ccl-program
908 (intern (format "%s-encoder" coding-system)))))
909 (if (and decoder encoder)
910 (aset coding-spec 4 (cons decoder encoder))
911 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
913 (setq coding-category 'coding-category-raw-text)))
915 (let ((plist (list 'coding-category coding-category
916 'alias-coding-systems (list coding-system))))
917 (if no-initial-designation
918 (plist-put plist 'no-initial-designation t))
920 (or (eq properties t)
921 (not (consp (car properties)))))
922 ;; In the old version, the arg PROPERTIES is a list to be
923 ;; set in PLIST as a value of property `safe-charsets'.
924 (setq properties (list (cons 'safe-charsets properties))))
925 ;; In the current version PROPERTIES is a property list.
926 ;; Reflect it into PLIST one by one while handling safe-chars
928 (let ((safe-charsets (cdr (assq 'safe-charsets properties)))
929 (safe-chars (cdr (assq 'safe-chars properties)))
932 ;; If only safe-charsets is specified, make a char-table from
933 ;; it, and store that char-table as the value of `safe-chars'.
934 (if (and (not safe-chars) safe-charsets)
936 (if (eq safe-charsets t)
938 (setq safe-chars (make-char-table 'safe-chars))
940 (setq charset (car safe-charsets)
941 safe-charsets (cdr safe-charsets))
942 (cond ((eq charset 'ascii)) ; just ignore
943 ((eq charset 'eight-bit-control)
946 (aset safe-chars i t)
948 ((eq charset 'eight-bit-graphic)
951 (aset safe-chars i t)
954 (aset safe-chars (make-char charset) t))))
955 (if accept-latin-extra-code
958 (if (aref latin-extra-code-table i)
959 (aset safe-chars i t))
961 (setq l (cons (cons 'safe-chars safe-chars) l))))
963 (setq prop (car (car l)) val (cdr (car l)) l (cdr l))
964 (if (eq prop 'safe-chars)
966 (if (and (symbolp val)
967 (get val 'translation-table))
968 (setq safe-chars (get val 'translation-table)))
969 (register-char-codings coding-system safe-chars)
970 (setq val safe-chars)))
971 (plist-put plist prop val)))
972 ;; The property `coding-category' may have been set differently
973 ;; through PROPERTIES.
974 (setq coding-category (plist-get plist 'coding-category))
975 (aset coding-spec coding-spec-plist-idx plist))
976 (put coding-system 'coding-system coding-spec)
977 (put coding-category 'coding-systems
978 (cons coding-system (get coding-category 'coding-systems))))
980 ;; Next, set a value of `eol-type' property.
982 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
983 ;; systems, each corresponds to a coding system for the detected
985 (setq eol-type (make-subsidiary-coding-system coding-system)))
987 (cond ((or (eq eol-type 'unix) (null eol-type))
993 ((or (and (vectorp eol-type)
994 (= (length eol-type) 3))
995 (and (numberp eol-type)
1000 (error "Invalid EOL-TYPE spec:%S" eol-type))))
1001 (put coding-system 'eol-type eol-type)
1003 ;; At last, register CODING-SYSTEM in `coding-system-list' and
1004 ;; `coding-system-alist'.
1005 (add-to-coding-system-list coding-system)
1006 (setq coding-system-alist (cons (list (symbol-name coding-system))
1007 coding-system-alist))
1009 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
1010 ;; XXX-with-esc variants.
1011 (let ((coding-category (coding-system-category coding-system)))
1012 (if (or (eq coding-category 'coding-category-iso-8-1)
1013 (eq coding-category 'coding-category-iso-8-2))
1014 (let ((esc (intern (concat (symbol-name coding-system) "-with-esc")))
1015 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system))
1016 (safe-charsets (assq 'safe-charsets properties))
1017 (mime-charset (assq 'mime-charset properties)))
1019 (setcdr safe-charsets t)
1020 (setq properties (cons (cons 'safe-charsets t) properties)))
1022 (setcdr mime-charset nil))
1023 (make-coding-system esc type mnemonic doc
1024 (if (listp (car flags))
1025 (cons (append (car flags) '(t)) (cdr flags))
1026 (cons (list (car flags) t) (cdr flags)))
1031 (defun define-coding-system-alias (alias coding-system)
1032 "Define ALIAS as an alias for coding system CODING-SYSTEM."
1033 (put alias 'coding-system (coding-system-spec coding-system))
1034 (nconc (coding-system-get alias 'alias-coding-systems) (list alias))
1035 (add-to-coding-system-list alias)
1036 (setq coding-system-alist (cons (list (symbol-name alias))
1037 coding-system-alist))
1038 (let ((eol-type (coding-system-eol-type coding-system)))
1039 (if (vectorp eol-type)
1040 (put alias 'eol-type (make-subsidiary-coding-system alias))
1041 (put alias 'eol-type eol-type))))
1043 (defun set-buffer-file-coding-system (coding-system &optional force)
1044 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1045 This means that when you save the buffer, it will be converted
1046 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
1047 use \\[list-coding-systems].
1049 If the buffer's previous file coding-system value specifies end-of-line
1050 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
1051 merged with the already-specified end-of-line conversion.
1053 If the buffer's previous file coding-system value specifies text
1054 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
1055 merged with the already-specified text conversion.
1057 However, if the optional prefix argument FORCE is non-nil, then
1058 CODING-SYSTEM is used exactly as specified.
1060 This marks the buffer modified so that the succeeding \\[save-buffer]
1061 surely saves the buffer with CODING-SYSTEM. From a program, if you
1062 don't want to mark the buffer modified, just set the variable
1063 `buffer-file-coding-system' directly."
1064 (interactive "zCoding system for visited file (default, nil): \nP")
1065 (check-coding-system coding-system)
1066 (if (and coding-system buffer-file-coding-system (null force))
1067 (let ((base (coding-system-base buffer-file-coding-system))
1068 (eol (coding-system-eol-type buffer-file-coding-system)))
1069 ;; If CODING-SYSTEM doesn't specify text conversion, merge
1070 ;; with that of buffer-file-coding-system.
1071 (if (eq (coding-system-base coding-system) 'undecided)
1072 (setq coding-system (coding-system-change-text-conversion
1073 coding-system base)))
1074 ;; If CODING-SYSTEM doesn't specify eol conversion, merge with
1075 ;; that of buffer-file-coding-system.
1076 (if (and (vectorp (coding-system-eol-type coding-system))
1077 (numberp eol) (>= eol 0) (<= eol 2))
1078 (setq coding-system (coding-system-change-eol-conversion
1079 coding-system eol)))))
1080 (setq buffer-file-coding-system coding-system)
1081 (set-buffer-modified-p t)
1082 (force-mode-line-update))
1084 (defvar default-terminal-coding-system nil
1085 "Default value for the terminal coding system.
1086 This is normally set according to the selected language environment.
1087 See also the command `set-terminal-coding-system'.")
1089 (defun set-terminal-coding-system (coding-system)
1090 "Set coding system of your terminal to CODING-SYSTEM.
1091 All text output to the terminal will be encoded
1092 with the specified coding system.
1093 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1094 The default is determined by the selected language environment
1095 or by the previous use of this command."
1097 (list (let ((default (if (and (not (terminal-coding-system))
1098 default-terminal-coding-system)
1099 default-terminal-coding-system)))
1101 (format "Coding system for terminal display (default, %s): "
1104 (if (and (not coding-system)
1105 (not (terminal-coding-system)))
1106 (setq coding-system default-terminal-coding-system))
1108 (setq default-terminal-coding-system coding-system))
1109 (set-terminal-coding-system-internal coding-system)
1110 (redraw-frame (selected-frame)))
1112 (defvar default-keyboard-coding-system nil
1113 "Default value of the keyboard coding system.
1114 This is normally set according to the selected language environment.
1115 See also the command `set-keyboard-coding-system'.")
1117 (defun set-keyboard-coding-system (coding-system)
1118 "Set coding system for keyboard input to CODING-SYSTEM.
1119 In addition, this command enables Encoded-kbd minor mode.
1120 \(If CODING-SYSTEM is nil, Encoded-kbd mode is turned off -- see
1121 `encoded-kbd-mode'.)
1122 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1123 The default is determined by the selected language environment
1124 or by the previous use of this command."
1126 (list (let ((default (if (and (not (keyboard-coding-system))
1127 default-keyboard-coding-system)
1128 default-keyboard-coding-system)))
1130 (format "Coding system for keyboard input (default, %s): "
1133 (if (and (not coding-system)
1134 (not (keyboard-coding-system)))
1135 (setq coding-system default-keyboard-coding-system))
1137 (setq default-keyboard-coding-system coding-system))
1138 (set-keyboard-coding-system-internal coding-system)
1139 (setq keyboard-coding-system coding-system)
1140 (encoded-kbd-mode (if coding-system 1 0)))
1142 (defcustom keyboard-coding-system nil
1143 "Specify coding system for keyboard input.
1144 If you set this on a terminal which can't distinguish Meta keys from
1145 8-bit characters, you will have to use ESC to type Meta characters.
1146 See Info node `Specify Coding' and Info node `Single-Byte Character Support'.
1148 Setting this variable directly does not take effect;
1149 use either M-x customize or \\[set-keyboard-coding-system]."
1150 :type '(coding-system :tag "Coding system")
1151 :link '(info-link "(emacs)Specify Coding")
1152 :link '(info-link "(emacs)Single-Byte Character Support")
1153 :set (lambda (symbol value)
1154 ;; Don't load encoded-kbd-mode unnecessarily.
1155 (if (or value (boundp 'encoded-kbd-mode))
1156 (set-keyboard-coding-system value)
1157 (set-default 'keyboard-coding-system nil))) ; must initialize
1162 (defun set-buffer-process-coding-system (decoding encoding)
1163 "Set coding systems for the process associated with the current buffer.
1164 DECODING is the coding system to be used to decode input from the process,
1165 ENCODING is the coding system to be used to encode output to the process.
1167 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
1169 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1170 (let ((proc (get-buffer-process (current-buffer))))
1172 (error "No process")
1173 (check-coding-system decoding)
1174 (check-coding-system encoding)
1175 (set-process-coding-system proc decoding encoding)))
1176 (force-mode-line-update))
1178 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
1180 (defun set-selection-coding-system (coding-system)
1181 "Make CODING-SYSTEM used for communicating with other X clients .
1182 When sending or receiving text via cut_buffer, selection, and clipboard,
1183 the text is encoded or decoded by CODING-SYSTEM."
1184 (interactive "zCoding system for X selection: ")
1185 (check-coding-system coding-system)
1186 (setq selection-coding-system coding-system))
1188 ;; Coding system lastly specified by the command
1189 ;; set-next-selection-coding-system.
1190 (defvar last-next-selection-coding-system nil)
1192 (defun set-next-selection-coding-system (coding-system)
1193 "Make CODING-SYSTEM used for the next communication with other X clients.
1194 This setting is effective for the next communication only."
1196 (list (read-coding-system
1197 (if last-next-selection-coding-system
1198 (format "Coding system for the next X selection (default, %S): "
1199 last-next-selection-coding-system)
1200 "Coding system for the next X selection: ")
1201 last-next-selection-coding-system)))
1203 (setq last-next-selection-coding-system coding-system)
1204 (setq coding-system last-next-selection-coding-system))
1205 (check-coding-system coding-system)
1207 (setq next-selection-coding-system coding-system))
1209 (defun set-coding-priority (arg)
1210 "Set priority of coding categories according to ARG.
1211 ARG is a list of coding categories ordered by priority."
1213 (current-list (copy-sequence coding-category-list)))
1214 ;; Check the validity of ARG while deleting coding categories in
1215 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
1216 ;; contains all coding categories.
1218 (if (or (null (get (car l) 'coding-category-index))
1219 (null (memq (car l) current-list)))
1220 (error "Invalid or duplicated element in argument: %s" arg))
1221 (setq current-list (delq (car l) current-list))
1223 ;; Update `coding-category-list' and return it.
1224 (setq coding-category-list (append arg current-list))
1225 (set-coding-priority-internal)))
1229 (defcustom auto-coding-alist
1230 '(("\\.\\(arc\\|zip\\|lzh\\|zoo\\|jar\\|tar\\|tgz\\)\\'" . no-conversion)
1231 ("\\.\\(gz\\|Z\\|bz\\|bz2\\|gpg\\)\\'" . no-conversion))
1232 "Alist of filename patterns vs corresponding coding systems.
1233 Each element looks like (REGEXP . CODING-SYSTEM).
1234 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1236 The settings in this alist take priority over `coding:' tags
1237 in the file (see the function `set-auto-coding')
1238 and the contents of `file-coding-system-alist'."
1241 :type '(repeat (cons (regexp :tag "File name regexp")
1242 (symbol :tag "Coding system"))))
1244 (defcustom auto-coding-regexp-alist
1245 '(("^BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion))
1246 "Alist of patterns vs corresponding coding systems.
1247 Each element looks like (REGEXP . CODING-SYSTEM).
1248 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1250 The settings in this alist take priority over `coding:' tags
1251 in the file (see the function `set-auto-coding')
1252 and the contents of `file-coding-system-alist'."
1255 :type '(repeat (cons (regexp :tag "Regexp")
1256 (symbol :tag "Coding system"))))
1258 (defvar set-auto-coding-for-load nil
1259 "Non-nil means look for `load-coding' property instead of `coding'.
1260 This is used for loading and byte-compiling Emacs Lisp files.")
1262 (defun auto-coding-alist-lookup (filename)
1263 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1264 (let ((alist auto-coding-alist)
1265 (case-fold-search (memq system-type '(vax-vms windows-nt ms-dos)))
1267 (while (and alist (not coding-system))
1268 (if (string-match (car (car alist)) filename)
1269 (setq coding-system (cdr (car alist)))
1270 (setq alist (cdr alist))))
1274 (defun auto-coding-from-file-contents (size)
1275 "Determine a coding system from the contents of the current buffer.
1276 The current buffer contains SIZE bytes starting at point.
1277 Value is either a coding system or nil."
1279 (let ((alist auto-coding-regexp-alist)
1281 (while (and alist (not coding-system))
1282 (let ((regexp (car (car alist))))
1283 (when (re-search-forward regexp (+ (point) size) t)
1284 (setq coding-system (cdr (car alist)))))
1285 (setq alist (cdr alist)))
1289 (defun set-auto-coding (filename size)
1290 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1291 These bytes should include at least the first 1k of the file
1292 and the last 3k of the file, but the middle may be omitted.
1294 It checks FILENAME against the variable `auto-coding-alist'. If
1295 FILENAME doesn't match any entries in the variable, it checks the
1296 contents of the current buffer following point against
1297 `auto-coding-regexp-alist'. If no match is found, it checks for a
1298 `coding:' tag in the first one or two lines following point. If no
1299 `coding:' tag is found, it checks for local variables list in the last
1300 3K bytes out of the SIZE bytes.
1302 The return value is the specified coding system,
1303 or nil if nothing specified.
1305 The variable `set-auto-coding-function' (which see) is set to this
1306 function by default."
1307 (or (auto-coding-alist-lookup filename)
1308 (auto-coding-from-file-contents size)
1309 (let* ((case-fold-search t)
1310 (head-start (point))
1311 (head-end (+ head-start (min size 1024)))
1312 (tail-start (+ head-start (max (- size 3072) 0)))
1313 (tail-end (+ head-start size))
1314 coding-system head-found tail-found pos)
1315 ;; Try a short cut by searching for the string "coding:"
1316 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1317 (setq head-found (or (search-forward "coding:" head-end t)
1318 (search-forward "unibyte:" head-end t)))
1319 (if (and head-found (> head-found tail-start))
1320 ;; Head and tail are overlapped.
1321 (setq tail-found head-found)
1322 (goto-char tail-start)
1323 (setq tail-found (or (search-forward "coding:" tail-end t)
1324 (search-forward "unibyte:" tail-end t))))
1326 ;; At first check the head.
1328 (goto-char head-start)
1329 (setq pos (re-search-forward "[\n\r]" head-end t))
1331 (= (char-after head-start) ?#)
1332 (= (char-after (1+ head-start)) ?!))
1333 ;; If the file begins with "#!" (exec interpreter magic),
1334 ;; look for coding frobs in the first two lines. You cannot
1335 ;; necessarily put them in the first line of such a file
1336 ;; without screwing up the interpreter invocation.
1337 (setq pos (search-forward "\n" head-end t)))
1338 (if pos (setq head-end pos))
1339 (when (< head-found head-end)
1340 (goto-char head-start)
1341 (when (and set-auto-coding-for-load
1343 "-\\*-\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1345 (setq coding-system 'raw-text))
1346 (when (and (not coding-system)
1348 "-\\*-\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1350 (setq coding-system (intern (match-string 2)))
1351 (or (coding-system-p coding-system)
1352 (setq coding-system nil)))))
1354 ;; If no coding: tag in the head, check the tail.
1355 (when (and tail-found (not coding-system))
1356 (goto-char tail-start)
1357 (search-forward "\n\^L" nil t)
1358 (if (re-search-forward
1359 "^\\(.*\\)[ \t]*Local Variables:[ \t]*\\(.*\\)$" tail-end t)
1360 ;; The prefix is what comes before "local variables:" in its
1361 ;; line. The suffix is what comes after "local variables:"
1363 (let* ((prefix (regexp-quote (match-string 1)))
1364 (suffix (regexp-quote (match-string 2)))
1368 ;; N.B. without the \n below, the regexp can
1370 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t\n]+\\)[ \t]*"
1375 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t\n]+\\)[ \t]*"
1378 (concat "^" prefix "[ \t]*End *:[ \t]*" suffix "$"))
1380 (re-search-forward re-end tail-end 'move)
1381 (setq tail-end (point))
1383 (when (and set-auto-coding-for-load
1384 (re-search-forward re-unibyte tail-end t))
1385 (setq coding-system 'raw-text))
1386 (when (and (not coding-system)
1387 (re-search-forward re-coding tail-end t))
1388 (setq coding-system (intern (match-string 1)))
1389 (or (coding-system-p coding-system)
1390 (setq coding-system nil))))))
1393 (setq set-auto-coding-function 'set-auto-coding)
1395 (defun after-insert-file-set-buffer-file-coding-system (inserted)
1396 "Set `buffer-file-coding-system' of current buffer after text is inserted."
1397 (if last-coding-system-used
1398 (let ((coding-system
1399 (find-new-buffer-file-coding-system last-coding-system-used))
1400 (modified-p (buffer-modified-p)))
1402 (set-buffer-file-coding-system coding-system t)
1403 (if (and enable-multibyte-characters
1404 (or (eq coding-system 'no-conversion)
1405 (eq (coding-system-type coding-system) 5))
1406 ;; If buffer was unmodified and the size is the
1407 ;; same as INSERTED, we must be visiting it.
1409 (= (buffer-size) inserted))
1410 ;; For coding systems no-conversion and raw-text...,
1411 ;; edit the buffer as unibyte.
1412 (let ((pos-byte (position-bytes (+ (point) inserted))))
1413 (set-buffer-multibyte nil)
1414 (setq inserted (- pos-byte (position-bytes (point))))))
1415 (set-buffer-modified-p modified-p))))
1418 (add-hook 'after-insert-file-functions
1419 'after-insert-file-set-buffer-file-coding-system)
1421 ;; The coding-spec and eol-type of coding-system returned is decided
1422 ;; independently in the following order.
1423 ;; 1. That of buffer-file-coding-system locally bound.
1424 ;; 2. That of CODING.
1426 (defun find-new-buffer-file-coding-system (coding)
1427 "Return a coding system for a buffer when a file of CODING is inserted.
1428 The local variable `buffer-file-coding-system' of the current buffer
1429 is set to the returned value.
1430 Return nil if there's no need to set `buffer-file-coding-system'."
1431 (let (local-coding local-eol
1432 found-coding found-eol
1435 ;; Nothing found about coding.
1438 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1439 ;; and LOCAL-CODING.
1440 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1441 (if (null (numberp local-eol))
1442 ;; But eol-type is not yet set.
1443 (setq local-eol nil))
1444 (if (and buffer-file-coding-system
1445 (not (eq (coding-system-type buffer-file-coding-system) t)))
1446 ;; This is not `undecided'.
1447 (setq local-coding (coding-system-base buffer-file-coding-system)))
1449 (if (and (local-variable-p 'buffer-file-coding-system)
1450 local-eol local-coding)
1451 ;; The current buffer has already set full coding-system, we
1452 ;; had better not change it.
1455 (setq found-eol (coding-system-eol-type coding))
1456 (if (null (numberp found-eol))
1457 ;; But eol-type is not found.
1458 ;; If EOL conversions are inhibited, force unix eol-type.
1459 (setq found-eol (if inhibit-eol-conversion 0)))
1460 (if (eq (coding-system-type coding) t)
1461 (setq found-coding 'undecided)
1462 (setq found-coding (coding-system-base coding)))
1464 (if (and (not found-eol) (eq found-coding 'undecided))
1465 ;; No valid coding information found.
1468 ;; Some coding information (eol or text) found.
1470 ;; The local setting takes precedence over the found one.
1471 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
1472 (or local-coding found-coding)
1473 (or found-coding local-coding)))
1474 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
1475 (or local-eol found-eol)
1476 (or found-eol local-eol)))
1478 (let ((eol-type (coding-system-eol-type new-coding)))
1479 (if (and (numberp new-eol) (vectorp eol-type))
1480 (aref eol-type new-eol)
1483 (defun modify-coding-system-alist (target-type regexp coding-system)
1484 "Modify one of look up tables for finding a coding system on I/O operation.
1485 There are three of such tables, `file-coding-system-alist',
1486 `process-coding-system-alist', and `network-coding-system-alist'.
1488 TARGET-TYPE specifies which of them to modify.
1489 If it is `file', it affects `file-coding-system-alist' (which see).
1490 If it is `process', it affects `process-coding-system-alist' (which see).
1491 If it is `network', it affects `network-coding-system-alist' (which see).
1493 REGEXP is a regular expression matching a target of I/O operation.
1494 The target is a file name if TARGET-TYPE is `file', a program name if
1495 TARGET-TYPE is `process', or a network service name or a port number
1496 to connect to if TARGET-TYPE is `network'.
1498 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1499 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1500 for decoding and encoding respectively,
1501 or a function symbol which, when called, returns such a cons cell."
1502 (or (memq target-type '(file process network))
1503 (error "Invalid target type: %s" target-type))
1504 (or (stringp regexp)
1505 (and (eq target-type 'network) (integerp regexp))
1506 (error "Invalid regular expression: %s" regexp))
1507 (if (symbolp coding-system)
1508 (if (not (fboundp coding-system))
1510 (check-coding-system coding-system)
1511 (setq coding-system (cons coding-system coding-system))))
1512 (check-coding-system (car coding-system))
1513 (check-coding-system (cdr coding-system)))
1514 (cond ((eq target-type 'file)
1515 (let ((slot (assoc regexp file-coding-system-alist)))
1517 (setcdr slot coding-system)
1518 (setq file-coding-system-alist
1519 (cons (cons regexp coding-system)
1520 file-coding-system-alist)))))
1521 ((eq target-type 'process)
1522 (let ((slot (assoc regexp process-coding-system-alist)))
1524 (setcdr slot coding-system)
1525 (setq process-coding-system-alist
1526 (cons (cons regexp coding-system)
1527 process-coding-system-alist)))))
1529 (let ((slot (assoc regexp network-coding-system-alist)))
1531 (setcdr slot coding-system)
1532 (setq network-coding-system-alist
1533 (cons (cons regexp coding-system)
1534 network-coding-system-alist)))))))
1536 (defun make-translation-table (&rest args)
1537 "Make a translation table from arguments.
1538 A translation table is a char table intended for character
1539 translation in CCL programs.
1541 Each argument is a list of elements of the form (FROM . TO), where FROM
1542 is a character to be translated to TO.
1544 FROM can be a generic character (see `make-char'). In this case, TO is
1545 a generic character containing the same number of characters, or an
1546 ordinary character. If FROM and TO are both generic characters, all
1547 characters belonging to FROM are translated to characters belonging to TO
1548 without changing their position code(s).
1550 The arguments and forms in each argument are processed in the given
1551 order, and if a previous form already translates TO to some other
1552 character, say TO-ALT, FROM is also translated to TO-ALT."
1553 (let ((table (make-char-table 'translation-table))
1556 (let ((elts (car args)))
1558 (let* ((from (car (car elts)))
1559 (from-i 0) ; degree of freedom of FROM
1560 (from-rev (nreverse (split-char from)))
1561 (to (cdr (car elts)))
1562 (to-i 0) ; degree of freedom of TO
1563 (to-rev (nreverse (split-char to))))
1564 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
1565 (while (eq (car from-rev) 0)
1566 (setq from-i (1+ from-i) from-rev (cdr from-rev)))
1567 (while (eq (car to-rev) 0)
1568 (setq to-i (1+ to-i) to-rev (cdr to-rev)))
1569 (if (and (/= from-i to-i) (/= to-i 0))
1570 (error "Invalid character pair (%d . %d)" from to))
1571 ;; If we have already translated TO to TO-ALT, FROM should
1572 ;; also be translated to TO-ALT. But, this is only if TO
1573 ;; is a generic character or TO-ALT is not a generic
1575 (let ((to-alt (aref table to)))
1577 (or (> to-i 0) (not (generic-char-p to-alt))))
1580 (set-char-table-default table from to)
1581 (aset table from to))
1582 ;; If we have already translated some chars to FROM, they
1583 ;; should also be translated to TO.
1584 (let ((l (assq from revlist)))
1591 (setq l (cdr l)) ))))
1592 ;; Now update REVLIST.
1593 (let ((l (assq to revlist)))
1595 (setcdr l (cons from (cdr l)))
1596 (setq revlist (cons (list to from) revlist)))))
1597 (setq elts (cdr elts))))
1598 (setq args (cdr args)))
1599 ;; Return TABLE just created.
1602 (defun make-translation-table-from-vector (vec)
1603 "Make translation table from decoding vector VEC.
1604 VEC is an array of 256 elements to map unibyte codes to multibyte characters.
1605 See also the variable `nonascii-translation-table'."
1606 (let ((table (make-char-table 'translation-table))
1607 (rev-table (make-char-table 'translation-table))
1611 (setq ch (aref vec i))
1614 (aset rev-table ch i))
1616 (set-char-table-extra-slot table 0 rev-table)
1619 (defun define-translation-table (symbol &rest args)
1620 "Define SYMBOL as the name of translation table made by ARGS.
1621 This sets up information so that the table can be used for
1622 translations in a CCL program.
1624 If the first element of ARGS is a char-table whose purpose is
1625 `translation-table', just define SYMBOL to name it. (Note that this
1626 function does not bind SYMBOL.)
1628 Any other ARGS should be suitable as arguments of the function
1629 `make-translation-table' (which see).
1631 This function sets properties `translation-table' and
1632 `translation-table-id' of SYMBOL to the created table itself and the
1633 identification number of the table respectively. It also registers
1634 the table in `translation-table-vector'."
1635 (let ((table (if (and (char-table-p (car args))
1636 (eq (char-table-subtype (car args))
1637 'translation-table))
1639 (apply 'make-translation-table args)))
1640 (len (length translation-table-vector))
1643 (put symbol 'translation-table table)
1646 (setq translation-table-vector
1647 (vconcat translation-table-vector (make-vector len nil))))
1648 (let ((slot (aref translation-table-vector id)))
1650 (eq (car slot) symbol))
1652 (aset translation-table-vector id (cons symbol table))
1654 (setq id (1+ id)))))
1655 (put symbol 'translation-table-id id)
1658 (put 'with-category-table 'lisp-indent-function 1)
1660 (defmacro with-category-table (category-table &rest body)
1661 `(let ((current-category-table (category-table)))
1662 (set-category-table ,category-table)
1665 (set-category-table current-category-table))))
1667 ;;; Initialize some variables.
1669 (put 'use-default-ascent 'char-table-extra-slots 0)
1670 (setq use-default-ascent (make-char-table 'use-default-ascent))
1671 (put 'ignore-relative-composition 'char-table-extra-slots 0)
1672 (setq ignore-relative-composition
1673 (make-char-table 'ignore-relative-composition))
1678 ;;; mule.el ends here