]> code.delx.au - gnu-emacs/blob - lisp/international/mule.el
(decode-char): Fix a typo. From Pavel Janik <Pavel@Janik.cz>.
[gnu-emacs] / lisp / international / mule.el
1 ;;; mule.el --- basic commands for mulitilingual environment
2
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Copyright (C) 2001 Free Software Foundation, Inc.
5 ;; Licensed to the Free Software Foundation.
6
7 ;; Keywords: mule, multilingual, character set, coding system
8
9 ;; This file is part of GNU Emacs.
10
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)
14 ;; any later version.
15
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.
20
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.
25
26 ;;; Code:
27
28 (defconst mule-version "5.0 (SAKAKI)" "\
29 Version number and name of this version of MULE (multilingual environment).")
30
31 (defconst mule-version-date "1999.12.7" "\
32 Distribution date of this version of MULE (multilingual environment).")
33
34 (defun load-with-code-conversion (fullname file &optional noerror nomessage)
35 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
36 The file contents are decoded before evaluation if necessary.
37 If optional second arg NOERROR is non-nil,
38 report no error if FILE doesn't exist.
39 Print messages at start and end of loading unless
40 optional third arg NOMESSAGE is non-nil.
41 Return t if file exists."
42 (if (null (file-readable-p fullname))
43 (and (null noerror)
44 (signal 'file-error (list "Cannot open load file" file)))
45 ;; Read file with code conversion, and then eval.
46 (let* ((buffer
47 ;; To avoid any autoloading, set default-major-mode to
48 ;; fundamental-mode.
49 ;; So that we don't get completely screwed if the
50 ;; file is encoded in some complicated character set,
51 ;; read it with real decoding, as a multibyte buffer,
52 ;; even if this is a --unibyte Emacs session.
53 (let ((default-major-mode 'fundamental-mode)
54 (default-enable-multibyte-characters t))
55 ;; We can't use `generate-new-buffer' because files.el
56 ;; is not yet loaded.
57 (get-buffer-create (generate-new-buffer-name " *load*"))))
58 (load-in-progress t)
59 (source (save-match-data (string-match "\\.el\\'" fullname))))
60 (unless nomessage
61 (if source
62 (message "Loading %s (source)..." file)
63 (message "Loading %s..." file)))
64 (when purify-flag
65 (setq preloaded-file-list (cons file preloaded-file-list)))
66 (unwind-protect
67 (let ((load-file-name fullname)
68 (set-auto-coding-for-load t)
69 (inhibit-file-name-operation nil))
70 (save-excursion
71 (set-buffer buffer)
72 (insert-file-contents fullname)
73 ;; If the loaded file was inserted with no-conversion or
74 ;; raw-text coding system, make the buffer unibyte.
75 ;; Otherwise, eval-buffer might try to interpret random
76 ;; binary junk as multibyte characters.
77 (if (and enable-multibyte-characters
78 (or (eq (coding-system-type last-coding-system-used) 5)
79 (eq last-coding-system-used 'no-conversion)))
80 (set-buffer-multibyte nil))
81 ;; Make `kill-buffer' quiet.
82 (set-buffer-modified-p nil))
83 ;; Have the original buffer current while we eval.
84 (eval-buffer buffer nil file
85 ;; If this Emacs is running with --unibyte,
86 ;; convert multibyte strings to unibyte
87 ;; after reading them.
88 ;; (not default-enable-multibyte-characters)
89 nil t
90 ))
91 (let (kill-buffer-hook kill-buffer-query-functions)
92 (kill-buffer buffer)))
93 (let ((hook (assoc file after-load-alist)))
94 (when hook
95 (mapcar (function eval) (cdr hook))))
96 (unless (or nomessage noninteractive)
97 (if source
98 (message "Loading %s (source)...done" file)
99 (message "Loading %s...done" file)))
100 t)))
101
102 ;; API (Application Program Interface) for charsets.
103
104 (defsubst charset-quoted-standard-p (obj)
105 "Return t if OBJ is a quoted symbol, and is the name of a standard charset."
106 (and (listp obj) (eq (car obj) 'quote)
107 (symbolp (car-safe (cdr obj)))
108 (let ((vector (get (car-safe (cdr obj)) 'charset)))
109 (and (vectorp vector)
110 (< (aref vector 0) 160)))))
111
112 (defsubst charsetp (object)
113 "T if OBJECT is a charset."
114 (and (symbolp object) (vectorp (get object 'charset))))
115
116 (defsubst charset-info (charset)
117 "Return a vector of information of CHARSET.
118 The elements of the vector are:
119 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
120 LEADING-CODE-BASE, LEADING-CODE-EXT,
121 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
122 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
123 PLIST,
124 where
125 CHARSET-ID (integer) is the identification number of the charset.
126 BYTES (integer) is the length of multi-byte form of a character in
127 the charset: one of 1, 2, 3, and 4.
128 DIMENSION (integer) is the number of bytes to represent a character of
129 the charset: 1 or 2.
130 CHARS (integer) is the number of characters in a dimension: 94 or 96.
131 WIDTH (integer) is the number of columns a character in the charset
132 occupies on the screen: one of 0, 1, and 2.
133 DIRECTION (integer) is the rendering direction of characters in the
134 charset when rendering. If 0, render from left to right, else
135 render from right to left.
136 LEADING-CODE-BASE (integer) is the base leading-code for the
137 charset.
138 LEADING-CODE-EXT (integer) is the extended leading-code for the
139 charset. All charsets of less than 0xA0 has the value 0.
140 ISO-FINAL-CHAR (character) is the final character of the
141 corresponding ISO 2022 charset. If the charset is not assigned
142 any final character, the value is -1.
143 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
144 while encoding to variants of ISO 2022 coding system, one of the
145 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
146 If the charset is not assigned any final character, the value is -1.
147 REVERSE-CHARSET (integer) is the charset which differs only in
148 LEFT-TO-RIGHT value from the charset. If there's no such a
149 charset, the value is -1.
150 SHORT-NAME (string) is the short name to refer to the charset.
151 LONG-NAME (string) is the long name to refer to the charset
152 DESCRIPTION (string) is the description string of the charset.
153 PLIST (property list) may contain any type of information a user
154 want to put and get by functions `put-charset-property' and
155 `get-charset-property' respectively."
156 (get charset 'charset))
157
158 ;; It is better not to use backquote in this file,
159 ;; because that makes a bootstrapping problem
160 ;; if you need to recompile all the Lisp files using interpreted code.
161
162 (defmacro charset-id (charset)
163 "Return charset identification number of CHARSET."
164 (if (charset-quoted-standard-p charset)
165 (aref (charset-info (nth 1 charset)) 0)
166 (list 'aref (list 'charset-info charset) 0)))
167
168 (defmacro charset-bytes (charset)
169 "Return bytes of CHARSET.
170 See the function `charset-info' for more detail."
171 (if (charset-quoted-standard-p charset)
172 (aref (charset-info (nth 1 charset)) 1)
173 (list 'aref (list 'charset-info charset) 1)))
174
175 (defmacro charset-dimension (charset)
176 "Return dimension of CHARSET.
177 See the function `charset-info' for more detail."
178 (if (charset-quoted-standard-p charset)
179 (aref (charset-info (nth 1 charset)) 2)
180 (list 'aref (list 'charset-info charset) 2)))
181
182 (defmacro charset-chars (charset)
183 "Return character numbers contained in a dimension of CHARSET.
184 See the function `charset-info' for more detail."
185 (if (charset-quoted-standard-p charset)
186 (aref (charset-info (nth 1 charset)) 3)
187 (list 'aref (list 'charset-info charset) 3)))
188
189 (defmacro charset-width (charset)
190 "Return width (how many column occupied on a screen) of CHARSET.
191 See the function `charset-info' for more detail."
192 (if (charset-quoted-standard-p charset)
193 (aref (charset-info (nth 1 charset)) 4)
194 (list 'aref (list 'charset-info charset) 4)))
195
196 (defmacro charset-direction (charset)
197 "Return direction of CHARSET.
198 See the function `charset-info' for more detail."
199 (if (charset-quoted-standard-p charset)
200 (aref (charset-info (nth 1 charset)) 5)
201 (list 'aref (list 'charset-info charset) 5)))
202
203 (defmacro charset-iso-final-char (charset)
204 "Return final char of CHARSET.
205 See the function `charset-info' for more detail."
206 (if (charset-quoted-standard-p charset)
207 (aref (charset-info (nth 1 charset)) 8)
208 (list 'aref (list 'charset-info charset) 8)))
209
210 (defmacro charset-iso-graphic-plane (charset)
211 "Return graphic plane of CHARSET.
212 See the function `charset-info' for more detail."
213 (if (charset-quoted-standard-p charset)
214 (aref (charset-info (nth 1 charset)) 9)
215 (list 'aref (list 'charset-info charset) 9)))
216
217 (defmacro charset-reverse-charset (charset)
218 "Return reverse charset of CHARSET.
219 See the function `charset-info' for more detail."
220 (if (charset-quoted-standard-p charset)
221 (aref (charset-info (nth 1 charset)) 10)
222 (list 'aref (list 'charset-info charset) 10)))
223
224 (defmacro charset-short-name (charset)
225 "Return short name of CHARSET.
226 See the function `charset-info' for more detail."
227 (if (charset-quoted-standard-p charset)
228 (aref (charset-info (nth 1 charset)) 11)
229 (list 'aref (list 'charset-info charset) 11)))
230
231 (defmacro charset-long-name (charset)
232 "Return long name of CHARSET.
233 See the function `charset-info' for more detail."
234 (if (charset-quoted-standard-p charset)
235 (aref (charset-info (nth 1 charset)) 12)
236 (list 'aref (list 'charset-info charset) 12)))
237
238 (defmacro charset-description (charset)
239 "Return description of CHARSET.
240 See the function `charset-info' for more detail."
241 (if (charset-quoted-standard-p charset)
242 (aref (charset-info (nth 1 charset)) 13)
243 (list 'aref (list 'charset-info charset) 13)))
244
245 (defmacro charset-plist (charset)
246 "Return list charset property of CHARSET.
247 See the function `charset-info' for more detail."
248 (list 'aref
249 (if (charset-quoted-standard-p charset)
250 (charset-info (nth 1 charset))
251 (list 'charset-info charset))
252 14))
253
254 (defun set-charset-plist (charset plist)
255 "Set CHARSET's property list to PLIST, and return PLIST."
256 (aset (charset-info charset) 14 plist))
257
258 (defun make-char (charset &optional code1 code2)
259 "Return a character of CHARSET whose position codes are CODE1 and CODE2.
260 CODE1 and CODE2 are optional, but if you don't supply
261 sufficient position codes, return a generic character which stands for
262 all characters or group of characters in the character set.
263 A generic character can be used to index a char table (e.g. syntax-table).
264
265 Such character sets as ascii, eight-bit-control, and eight-bit-graphic
266 don't have corresponding generic characters. If CHARSET is one of
267 them and you don't supply CODE1, return the character of the smallest
268 code in CHARSET.
269
270 If CODE1 or CODE2 are invalid (out of range), this function signals an error."
271 (make-char-internal (charset-id charset) code1 code2))
272
273 (put 'make-char 'byte-compile
274 (function
275 (lambda (form)
276 (let ((charset (nth 1 form)))
277 (if (charset-quoted-standard-p charset)
278 (byte-compile-normal-call
279 (cons 'make-char-internal
280 (cons (charset-id (nth 1 charset)) (nthcdr 2 form))))
281 (byte-compile-normal-call
282 (cons 'make-char-internal
283 (cons (list 'charset-id charset) (nthcdr 2 form)))))))))
284
285 (defun charset-list ()
286 "Return list of charsets ever defined.
287
288 This function is provided for backward compatibility.
289 Now we have the variable `charset-list'."
290 charset-list)
291
292 (defsubst generic-char-p (char)
293 "Return t if and only if CHAR is a generic character.
294 See also the documentation of `make-char'."
295 (and (>= char 0400)
296 (let ((l (split-char char)))
297 (and (or (= (nth 1 l) 0) (eq (nth 2 l) 0))
298 (not (eq (car l) 'composition))))))
299
300 (defun decode-char (ccs code-point &optional restriction)
301 "Return character specified by coded character set CCS and CODE-POINT in it.
302 Return nil if such a character is not supported.
303 Currently the only supported coded character set is `ucs' (ISO/IEC
304 10646: Universal Multi-Octet Coded Character Set).
305
306 Optional argument RESTRICTION specifies a way to map the pair of CCS
307 and CODE-POINT to a character. Currently not supported and just ignored."
308 (cond ((eq ccs 'ucs)
309 (cond ((< code-point 160)
310 code-point)
311 ((< code-point 256)
312 (make-char 'latin-iso8859-1 code-point))
313 ((< code-point #x2500)
314 (setq code-point (- code-point #x0100))
315 (make-char 'mule-unicode-0100-24ff
316 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
317 ((< code-point #x3400)
318 (setq code-point (- code-point #x2500))
319 (make-char 'mule-unicode-2500-33ff
320 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
321 ((and (>= code-point #xe000) (< code-point #x10000))
322 (setq code-point (- code-point #xe000))
323 (make-char 'mule-unicode-e000-ffff
324 (+ (/ code-point 96) 32) (+ (% code-point 96) 32)))
325 ))))
326
327 (defun encode-char (char ccs &optional restriction)
328 "Return code-point in coded character set CCS that corresponds to CHAR.
329 Return nil if CHAR is not included in CCS.
330 Currently the only supported coded character set is `ucs' (ISO/IEC
331 10646: Universal Multi-Octet Coded Character Set).
332
333 CHAR should be in one of these charsets:
334 ascii, latin-iso8859-1, mule-unicode-0100-24ff, mule-unicode-2500-33ff,
335 mule-unicode-e000-ffff, eight-bit-control
336 Otherwise, return nil.
337
338 Optional argument RESTRICTION specifies a way to map CHAR to a
339 code-point in CCS. Currently not supported and just ignored."
340 (let* ((split (split-char char))
341 (charset (car split)))
342 (cond ((eq ccs 'ucs)
343 (cond ((eq charset 'ascii)
344 char)
345 ((eq charset 'latin-iso8859-1)
346 (+ (nth 1 split) 128))
347 ((eq charset 'mule-unicode-0100-24ff)
348 (+ #x0100 (+ (* (- (nth 1 split) 32) 96)
349 (- (nth 2 split) 32))))
350 ((eq charset 'mule-unicode-2500-33ff)
351 (+ #x2500 (+ (* (- (nth 1 split) 32) 96)
352 (- (nth 2 split) 32))))
353 ((eq charset 'mule-unicode-e000-ffff)
354 (+ #xe000 (+ (* (- (nth 1 split) 32) 96)
355 (- (nth 2 split) 32))))
356 ((eq charset 'eight-bit-control)
357 char))))))
358
359 \f
360 ;; Coding system stuff
361
362 ;; Coding system is a symbol that has the property `coding-system'.
363 ;;
364 ;; The value of the property `coding-system' is a vector of the
365 ;; following format:
366 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
367 ;; We call this vector as coding-spec. See comments in src/coding.c
368 ;; for more detail.
369
370 (defconst coding-spec-type-idx 0)
371 (defconst coding-spec-mnemonic-idx 1)
372 (defconst coding-spec-doc-string-idx 2)
373 (defconst coding-spec-plist-idx 3)
374 (defconst coding-spec-flags-idx 4)
375
376 ;; PLIST is a property list of a coding system. To share PLIST among
377 ;; alias coding systems, a coding system has PLIST in coding-spec
378 ;; instead of having it in normal property list of Lisp symbol.
379 ;; Here's a list of coding system properties currently being used.
380 ;;
381 ;; o coding-category
382 ;;
383 ;; The value is a coding category the coding system belongs to. The
384 ;; function `make-coding-system' sets this value automatically
385 ;; unless its argument PROPERTIES specifies this property.
386 ;;
387 ;; o alias-coding-systems
388 ;;
389 ;; The value is a list of coding systems of the same alias group. The
390 ;; first element is the coding system made at first, which we call as
391 ;; `base coding system'. The function `make-coding-system' sets this
392 ;; value automatically and `define-coding-system-alias' updates it.
393 ;;
394 ;; See the documentation of make-coding-system for the meanings of the
395 ;; following properties.
396 ;;
397 ;; o post-read-conversion
398 ;; o pre-write-conversion
399 ;; o translation-table-for-decode
400 ;; o translation-table-for-encode
401 ;; o safe-chars
402 ;; o safe-charsets
403 ;; o mime-charset
404 ;; o valid-codes (meaningful only for a coding system based on CCL)
405
406
407 (defsubst coding-system-spec (coding-system)
408 "Return coding-spec of CODING-SYSTEM."
409 (get (check-coding-system coding-system) 'coding-system))
410
411 (defun coding-system-type (coding-system)
412 "Return the coding type of CODING-SYSTEM.
413 A coding type is an integer value indicating the encoding method
414 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
415 (aref (coding-system-spec coding-system) coding-spec-type-idx))
416
417 (defun coding-system-mnemonic (coding-system)
418 "Return the mnemonic character of CODING-SYSTEM.
419 The mnemonic character of a coding system is used in mode line
420 to indicate the coding system. If the arg is nil, return ?-."
421 (let ((spec (coding-system-spec coding-system)))
422 (if spec (aref spec coding-spec-mnemonic-idx) ?-)))
423
424 (defun coding-system-doc-string (coding-system)
425 "Return the documentation string for CODING-SYSTEM."
426 (aref (coding-system-spec coding-system) coding-spec-doc-string-idx))
427
428 (defun coding-system-plist (coding-system)
429 "Return the property list of CODING-SYSTEM."
430 (aref (coding-system-spec coding-system) coding-spec-plist-idx))
431
432 (defun coding-system-flags (coding-system)
433 "Return `flags' of CODING-SYSTEM.
434 A `flags' of a coding system is a vector of length 32 indicating detailed
435 information of a coding system. See the function `make-coding-system'
436 for more detail."
437 (aref (coding-system-spec coding-system) coding-spec-flags-idx))
438
439 (defun coding-system-get (coding-system prop)
440 "Extract a value from CODING-SYSTEM's property list for property PROP."
441 (plist-get (coding-system-plist coding-system) prop))
442
443 (defun coding-system-put (coding-system prop val)
444 "Change value in CODING-SYSTEM's property list PROP to VAL."
445 (let ((plist (coding-system-plist coding-system)))
446 (if plist
447 (plist-put plist prop val)
448 (aset (coding-system-spec coding-system) coding-spec-plist-idx
449 (list prop val)))))
450
451 (defun coding-system-category (coding-system)
452 "Return the coding category of CODING-SYSTEM.
453 See also `coding-category-list'."
454 (coding-system-get coding-system 'coding-category))
455
456 (defun coding-system-base (coding-system)
457 "Return the base coding system of CODING-SYSTEM.
458 A base coding system is what made by `make-coding-system'.
459 Any alias nor subsidiary coding systems are not base coding system."
460 (car (coding-system-get coding-system 'alias-coding-systems)))
461
462 (defalias 'coding-system-parent 'coding-system-base)
463 (make-obsolete 'coding-system-parent 'coding-system-base "20.3")
464
465 ;; Coding system also has a property `eol-type'.
466 ;;
467 ;; This property indicates how the coding system handles end-of-line
468 ;; format. The value is integer 0, 1, 2, or a vector of three coding
469 ;; systems. Each integer value 0, 1, and 2 indicates the format of
470 ;; end-of-line LF, CRLF, and CR respectively. A vector value
471 ;; indicates that the format of end-of-line should be detected
472 ;; automatically. Nth element of the vector is the subsidiary coding
473 ;; system whose `eol-type' property is N.
474
475 (defun coding-system-eol-type (coding-system)
476 "Return eol-type of CODING-SYSTEM.
477 An eol-type is integer 0, 1, 2, or a vector of coding systems.
478
479 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
480 CRLF, and CR respectively.
481
482 A vector value indicates that a format of end-of-line should be
483 detected automatically. Nth element of the vector is the subsidiary
484 coding system whose eol-type is N."
485 (get coding-system 'eol-type))
486
487 (defun coding-system-lessp (x y)
488 (cond ((eq x 'no-conversion) t)
489 ((eq y 'no-conversion) nil)
490 ((eq x 'emacs-mule) t)
491 ((eq y 'emacs-mule) nil)
492 ((eq x 'undecided) t)
493 ((eq y 'undecided) nil)
494 (t (let ((c1 (coding-system-mnemonic x))
495 (c2 (coding-system-mnemonic y)))
496 (or (< (downcase c1) (downcase c2))
497 (and (not (> (downcase c1) (downcase c2)))
498 (< c1 c2)))))))
499
500 (defun add-to-coding-system-list (coding-system)
501 "Add CODING-SYSTEM to `coding-system-list' while keeping it sorted."
502 (if (or (null coding-system-list)
503 (coding-system-lessp coding-system (car coding-system-list)))
504 (setq coding-system-list (cons coding-system coding-system-list))
505 (let ((len (length coding-system-list))
506 mid (tem coding-system-list))
507 (while (> len 1)
508 (setq mid (nthcdr (/ len 2) tem))
509 (if (coding-system-lessp (car mid) coding-system)
510 (setq tem mid
511 len (- len (/ len 2)))
512 (setq len (/ len 2))))
513 (setcdr tem (cons coding-system (cdr tem))))))
514
515 (defun coding-system-list (&optional base-only)
516 "Return a list of all existing non-subsidiary coding systems.
517 If optional arg BASE-ONLY is non-nil, only base coding systems are listed.
518 The value doesn't include subsidiary coding systems which are what
519 made from bases and aliases automatically for various end-of-line
520 formats (e.g. iso-latin-1-unix, koi8-r-dos)."
521 (let* ((codings (copy-sequence coding-system-list))
522 (tail (cons nil codings)))
523 ;; Remove subsidiary coding systems (eol variants) and alias
524 ;; coding systems (if necessary).
525 (while (cdr tail)
526 (let* ((coding (car (cdr tail)))
527 (aliases (coding-system-get coding 'alias-coding-systems)))
528 (if (or
529 ;; CODING is an eol variant if not in ALIASES.
530 (not (memq coding aliases))
531 ;; CODING is an alias if it is not car of ALIASES.
532 (and base-only (not (eq coding (car aliases)))))
533 (setcdr tail (cdr (cdr tail)))
534 (setq tail (cdr tail)))))
535 codings))
536
537 (defun register-char-codings (coding-system safe-chars)
538 (let ((general (char-table-extra-slot char-coding-system-table 0)))
539 (if (eq safe-chars t)
540 (or (memq coding-system general)
541 (set-char-table-extra-slot char-coding-system-table 0
542 (cons coding-system general)))
543 (map-char-table
544 (function
545 (lambda (key val)
546 (if (and (>= key 128) val)
547 (let ((codings (aref char-coding-system-table key)))
548 (or (memq coding-system codings)
549 (aset char-coding-system-table key
550 (cons coding-system codings)))))))
551 safe-chars))))
552
553
554 (defun make-subsidiary-coding-system (coding-system)
555 "Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM."
556 (let ((coding-spec (coding-system-spec coding-system))
557 (subsidiaries (vector (intern (format "%s-unix" coding-system))
558 (intern (format "%s-dos" coding-system))
559 (intern (format "%s-mac" coding-system))))
560 (i 0)
561 temp)
562 (while (< i 3)
563 (put (aref subsidiaries i) 'coding-system coding-spec)
564 (put (aref subsidiaries i) 'eol-type i)
565 (add-to-coding-system-list (aref subsidiaries i))
566 (setq coding-system-alist
567 (cons (list (symbol-name (aref subsidiaries i)))
568 coding-system-alist))
569 (setq i (1+ i)))
570 subsidiaries))
571
572 (defun transform-make-coding-system-args (name type &optional doc-string props)
573 "For internal use only.
574 Transform XEmacs style args for `make-coding-system' to Emacs style.
575 Value is a list of transformed arguments."
576 (let ((mnemonic (string-to-char (or (plist-get props 'mnemonic) "?")))
577 (eol-type (plist-get props 'eol-type))
578 properties tmp)
579 (cond
580 ((eq eol-type 'lf) (setq eol-type 'unix))
581 ((eq eol-type 'crlf) (setq eol-type 'dos))
582 ((eq eol-type 'cr) (setq eol-type 'mac)))
583 (if (setq tmp (plist-get props 'post-read-conversion))
584 (setq properties (plist-put properties 'post-read-conversion tmp)))
585 (if (setq tmp (plist-get props 'pre-write-conversion))
586 (setq properties (plist-put properties 'pre-write-conversion tmp)))
587 (cond
588 ((eq type 'shift-jis)
589 `(,name 1 ,mnemonic ,doc-string () ,properties ,eol-type))
590 ((eq type 'iso2022) ; This is not perfect.
591 (if (plist-get props 'escape-quoted)
592 (error "escape-quoted is not supported: %S"
593 `(,name ,type ,doc-string ,props)))
594 (let ((g0 (plist-get props 'charset-g0))
595 (g1 (plist-get props 'charset-g1))
596 (g2 (plist-get props 'charset-g2))
597 (g3 (plist-get props 'charset-g3))
598 (use-roman
599 (and
600 (eq (cadr (assoc 'latin-jisx0201
601 (plist-get props 'input-charset-conversion)))
602 'ascii)
603 (eq (cadr (assoc 'ascii
604 (plist-get props 'output-charset-conversion)))
605 'latin-jisx0201)))
606 (use-oldjis
607 (and
608 (eq (cadr (assoc 'japanese-jisx0208-1978
609 (plist-get props 'input-charset-conversion)))
610 'japanese-jisx0208)
611 (eq (cadr (assoc 'japanese-jisx0208
612 (plist-get props 'output-charset-conversion)))
613 'japanese-jisx0208-1978))))
614 (if (charsetp g0)
615 (if (plist-get props 'force-g0-on-output)
616 (setq g0 `(nil ,g0))
617 (setq g0 `(,g0 t))))
618 (if (charsetp g1)
619 (if (plist-get props 'force-g1-on-output)
620 (setq g1 `(nil ,g1))
621 (setq g1 `(,g1 t))))
622 (if (charsetp g2)
623 (if (plist-get props 'force-g2-on-output)
624 (setq g2 `(nil ,g2))
625 (setq g2 `(,g2 t))))
626 (if (charsetp g3)
627 (if (plist-get props 'force-g3-on-output)
628 (setq g3 `(nil ,g3))
629 (setq g3 `(,g3 t))))
630 `(,name 2 ,mnemonic ,doc-string
631 (,g0 ,g1 ,g2 ,g3
632 ,(plist-get props 'short)
633 ,(not (plist-get props 'no-ascii-eol))
634 ,(not (plist-get props 'no-ascii-cntl))
635 ,(plist-get props 'seven)
636 t
637 ,(not (plist-get props 'lock-shift))
638 ,use-roman
639 ,use-oldjis
640 ,(plist-get props 'no-iso6429)
641 nil nil nil nil)
642 ,properties ,eol-type)))
643 ((eq type 'big5)
644 `(,name 3 ,mnemonic ,doc-string () ,properties ,eol-type))
645 ((eq type 'ccl)
646 `(,name 4 ,mnemonic ,doc-string
647 (,(plist-get props 'decode) . ,(plist-get props 'encode))
648 ,properties ,eol-type))
649 (t
650 (error "unsupported XEmacs style make-coding-style arguments: %S"
651 `(,name ,type ,doc-string ,props))))))
652
653 (defun make-coding-system (coding-system type mnemonic doc-string
654 &optional
655 flags
656 properties
657 eol-type)
658 "Define a new coding system CODING-SYSTEM (symbol).
659 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
660 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
661 in the following format:
662 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
663
664 TYPE is an integer value indicating the type of the coding system as follows:
665 0: Emacs internal format,
666 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PC,
667 2: ISO-2022 including many variants,
668 3: Big5 used mainly on Chinese PC,
669 4: private, CCL programs provide encoding/decoding algorithm,
670 5: Raw-text, which means that text contains random 8-bit codes.
671
672 MNEMONIC is a character to be displayed on mode line for the coding system.
673
674 DOC-STRING is a documentation string for the coding system.
675
676 FLAGS specifies more detailed information of the coding system as follows:
677
678 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
679 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
680 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
681 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
682 SAFE, ACCEPT-LATIN-EXTRA-CODE.
683 CHARSETn are character sets initially designated to Gn graphic registers.
684 If CHARSETn is nil, Gn is never used.
685 If CHARSETn is t, Gn can be used but nothing designated initially.
686 If CHARSETn is a list of character sets, those character sets are
687 designated to Gn on output, but nothing designated to Gn initially.
688 But, character set `ascii' can be designated only to G0.
689 SHORT-FORM non-nil means use short designation sequence on output.
690 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
691 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
692 SPACE on output.
693 SEVEN non-nil means use 7-bit code only on output.
694 LOCKING-SHIFT non-nil means use locking-shift.
695 SINGLE-SHIFT non-nil means use single-shift.
696 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
697 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
698 NO-ISO6429 non-nil means not use ISO6429's direction specification.
699 INIT-BOL non-nil means any designation state is assumed to be reset
700 to initial at each beginning of line on output.
701 DESIGNATION-BOL non-nil means designation sequences should be placed
702 at beginning of line on output.
703 SAFE non-nil means convert unsafe characters to `?' on output.
704 Characters not specified in the property `safe-charsets' nor
705 `safe-chars' are unsafe.
706 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
707 a code specified in `latin-extra-code-table' (which see) as a valid
708 code of the coding system.
709
710 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
711 decoding and encoding. CCL programs should be specified by their
712 symbols.
713
714 PROPERTIES is an alist of properties vs the corresponding values. The
715 following properties are recognized:
716
717 o post-read-conversion
718
719 The value is a function to call after some text is inserted and
720 decoded by the coding system itself and before any functions in
721 `after-insert-functions' are called. The argument of this
722 function is the same as for a function in
723 `after-insert-file-functions', i.e. LENGTH of the text inserted,
724 with point at the head of the text to be decoded.
725
726 o pre-write-conversion
727
728 The value is a function to call after all functions in
729 `write-region-annotate-functions' and `buffer-file-format' are
730 called, and before the text is encoded by the coding system itself.
731 The arguments to this function are the same as those of a function
732 in `write-region-annotate-functions', i.e. FROM and TO, specifying
733 a region of text.
734
735 o translation-table-for-decode
736
737 The value is a translation table to be applied on decoding. See
738 the function `make-translation-table' for the format of translation
739 table. This is not applicable to type 4 (CCL-based) coding systems.
740
741 o translation-table-for-encode
742
743 The value is a translation table to be applied on encoding. This is
744 not applicable to type 4 (CCL-based) coding systems.
745
746 o safe-chars
747
748 The value is a char table. If a character has non-nil value in it,
749 the character is safely supported by the coding system. This
750 overrides the specification of safe-charsets.
751
752 o safe-charsets
753
754 The value is a list of charsets safely supported by the coding
755 system. The value t means that all charsets Emacs handles are
756 supported. Even if some charset is not in this list, it doesn't
757 mean that the charset can't be encoded in the coding system;
758 it just means that some other receiver of text encoded
759 in the coding system won't be able to handle that charset.
760
761 o mime-charset
762
763 The value is a symbol of which name is `MIME-charset' parameter of
764 the coding system.
765
766 o valid-codes (meaningful only for a coding system based on CCL)
767
768 The value is a list to indicate valid byte ranges of the encoded
769 file. Each element of the list is an integer or a cons of integer.
770 In the former case, the integer value is a valid byte code. In the
771 latter case, the integers specifies the range of valid byte codes.
772
773 These properties are set in PLIST, a property list. This function
774 also sets properties `coding-category' and `alias-coding-systems'
775 automatically.
776
777 EOL-TYPE specifies the EOL type of the coding-system in one of the
778 following formats:
779
780 o symbol (unix, dos, or mac)
781
782 The symbol `unix' means Unix-like EOL (LF), `dos' means
783 DOS-like EOL (CRLF), and `mac' means MAC-like EOL (CR).
784
785 o number (0, 1, or 2)
786
787 The number 0, 1, and 2 mean UNIX, DOS, and MAC-like EOL
788 respectively.
789
790 o vector of coding-systems of length 3
791
792 The EOL type is detected automatically for the coding system.
793 And, according to the detected EOL type, one of the coding
794 systems in the vector is selected. Elements of the vector
795 corresponds to Unix-like EOL, DOS-like EOL, and Mac-like EOL
796 in this order.
797
798 Kludgy features for backward compatibility:
799
800 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
801 treated as a compiled CCL code.
802
803 2. If PROPERTIES is just a list of character sets, the list is set as
804 a value of `safe-charsets' in PLIST."
805
806 ;; For compatiblity with XEmacs, we check the type of TYPE. If it
807 ;; is a symbol, perhaps, this function is called with XEmacs-style
808 ;; arguments. Here, try to transform that kind of arguments to
809 ;; Emacs style.
810 (if (symbolp type)
811 (let ((args (transform-make-coding-system-args coding-system type
812 mnemonic doc-string)))
813 (setq coding-system (car args)
814 type (nth 1 args)
815 mnemonic (nth 2 args)
816 doc-string (nth 3 args)
817 flags (nth 4 args)
818 properties (nth 5 args)
819 eol-type (nth 6 args))))
820
821 ;; Set a value of `coding-system' property.
822 (let ((coding-spec (make-vector 5 nil))
823 (no-initial-designation t)
824 (no-alternative-designation t)
825 (accept-latin-extra-code nil)
826 coding-category)
827 (if (or (not (integerp type)) (< type 0) (> type 5))
828 (error "TYPE argument must be 0..5"))
829 (if (or (not (integerp mnemonic)) (<= mnemonic ? ) (> mnemonic 127))
830 (error "MNEMONIC argument must be an ASCII printable character"))
831 (aset coding-spec coding-spec-type-idx type)
832 (aset coding-spec coding-spec-mnemonic-idx mnemonic)
833 (aset coding-spec coding-spec-doc-string-idx
834 (purecopy (if (stringp doc-string) doc-string "")))
835 (cond ((= type 0)
836 (setq coding-category 'coding-category-emacs-mule))
837 ((= type 1)
838 (setq coding-category 'coding-category-sjis))
839 ((= type 2) ; ISO2022
840 (let ((i 0)
841 (vec (make-vector 32 nil))
842 (g1-designation nil)
843 (fl flags))
844 (while (< i 4)
845 (let ((charset (car fl)))
846 (if (and no-initial-designation
847 (> i 0)
848 (or (charsetp charset)
849 (and (consp charset)
850 (charsetp (car charset)))))
851 (setq no-initial-designation nil))
852 (if (charsetp charset)
853 (if (= i 1) (setq g1-designation charset))
854 (if (consp charset)
855 (let ((tail charset)
856 elt)
857 (while tail
858 (setq elt (car tail))
859 (if (eq elt t)
860 (setq no-alternative-designation nil)
861 (if (and elt (not (charsetp elt)))
862 (error "Invalid charset: %s" elt)))
863 (setq tail (cdr tail)))
864 (setq g1-designation (car charset)))
865 (if charset
866 (if (eq charset t)
867 (setq no-alternative-designation nil)
868 (error "Invalid charset: %s" charset)))))
869 (aset vec i charset))
870 (setq fl (cdr fl) i (1+ i)))
871 (while (and (< i 32) fl)
872 (aset vec i (car fl))
873 (if (and (= i 16) ; ACCEPT-LATIN-EXTRA-CODE
874 (car fl))
875 (setq accept-latin-extra-code t))
876 (setq fl (cdr fl) i (1+ i)))
877 (aset coding-spec 4 vec)
878 (setq coding-category
879 (if (aref vec 8) ; Use locking-shift.
880 (or (and (aref vec 7) 'coding-category-iso-7-else)
881 'coding-category-iso-8-else)
882 (if (aref vec 7) ; 7-bit only.
883 (if (aref vec 9) ; Use single-shift.
884 'coding-category-iso-7-else
885 (if no-alternative-designation
886 'coding-category-iso-7-tight
887 'coding-category-iso-7))
888 (if (or no-initial-designation
889 (not no-alternative-designation))
890 'coding-category-iso-8-else
891 (if (and (charsetp g1-designation)
892 (= (charset-dimension g1-designation) 2))
893 'coding-category-iso-8-2
894 'coding-category-iso-8-1)))))))
895 ((= type 3)
896 (setq coding-category 'coding-category-big5))
897 ((= type 4) ; private
898 (setq coding-category 'coding-category-ccl)
899 (if (not (consp flags))
900 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
901 (let ((decoder (check-ccl-program
902 (car flags)
903 (intern (format "%s-decoder" coding-system))))
904 (encoder (check-ccl-program
905 (cdr flags)
906 (intern (format "%s-encoder" coding-system)))))
907 (if (and decoder encoder)
908 (aset coding-spec 4 (cons decoder encoder))
909 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
910 (t ; i.e. (= type 5)
911 (setq coding-category 'coding-category-raw-text)))
912
913 (let ((plist (list 'coding-category coding-category
914 'alias-coding-systems (list coding-system))))
915 (if no-initial-designation
916 (plist-put plist 'no-initial-designation t))
917 (if (and properties
918 (or (eq properties t)
919 (not (consp (car properties)))))
920 ;; In the old version, the arg PROPERTIES is a list to be
921 ;; set in PLIST as a value of property `safe-charsets'.
922 (setq properties (list (cons 'safe-charsets properties))))
923 ;; In the current version PROPERTIES is a property list.
924 ;; Reflect it into PLIST one by one while handling safe-chars
925 ;; specially.
926 (let ((safe-charsets (cdr (assq 'safe-charsets properties)))
927 (safe-chars (cdr (assq 'safe-chars properties)))
928 (l properties)
929 prop val)
930 ;; If only safe-charsets is specified, make a char-table from
931 ;; it, and store that char-table as the value of `safe-chars'.
932 (if (and (not safe-chars) safe-charsets)
933 (let (charset)
934 (if (eq safe-charsets t)
935 (setq safe-chars t)
936 (setq safe-chars (make-char-table 'safe-chars))
937 (while safe-charsets
938 (setq charset (car safe-charsets)
939 safe-charsets (cdr safe-charsets))
940 (cond ((eq charset 'ascii)) ; just ignore
941 ((eq charset 'eight-bit-control)
942 (let ((i 128))
943 (while (< i 160)
944 (aset safe-chars i t)
945 (setq i (1+ i)))))
946 ((eq charset 'eight-bit-graphic)
947 (let ((i 160))
948 (while (< i 256)
949 (aset safe-chars i t)
950 (setq i (1+ i)))))
951 (t
952 (aset safe-chars (make-char charset) t))))
953 (if accept-latin-extra-code
954 (let ((i 128))
955 (while (< i 160)
956 (if (aref latin-extra-code-table i)
957 (aset safe-chars i t))
958 (setq i (1+ i))))))
959 (setq l (cons (cons 'safe-chars safe-chars) l))))
960 (while l
961 (setq prop (car (car l)) val (cdr (car l)) l (cdr l))
962 (if (eq prop 'safe-chars)
963 (progn
964 (if (and (symbolp val)
965 (get val 'translation-table))
966 (setq safe-chars (get val 'translation-table)))
967 (register-char-codings coding-system safe-chars)
968 (setq val safe-chars)))
969 (plist-put plist prop val)))
970 ;; The property `coding-category' may have been set differently
971 ;; through PROPERTIES.
972 (setq coding-category (plist-get plist 'coding-category))
973 (aset coding-spec coding-spec-plist-idx plist))
974 (put coding-system 'coding-system coding-spec)
975 (put coding-category 'coding-systems
976 (cons coding-system (get coding-category 'coding-systems))))
977
978 ;; Next, set a value of `eol-type' property.
979 (if (not eol-type)
980 ;; If EOL-TYPE is nil, set a vector of subsidiary coding
981 ;; systems, each corresponds to a coding system for the detected
982 ;; EOL format.
983 (setq eol-type (make-subsidiary-coding-system coding-system)))
984 (setq eol-type
985 (cond ((or (eq eol-type 'unix) (null eol-type))
986 0)
987 ((eq eol-type 'dos)
988 1)
989 ((eq eol-type 'mac)
990 2)
991 ((or (and (vectorp eol-type)
992 (= (length eol-type) 3))
993 (and (numberp eol-type)
994 (and (>= eol-type 0)
995 (<= eol-type 2))))
996 eol-type)
997 (t
998 (error "Invalid EOL-TYPE spec:%S" eol-type))))
999 (put coding-system 'eol-type eol-type)
1000
1001 ;; At last, register CODING-SYSTEM in `coding-system-list' and
1002 ;; `coding-system-alist'.
1003 (add-to-coding-system-list coding-system)
1004 (setq coding-system-alist (cons (list (symbol-name coding-system))
1005 coding-system-alist))
1006
1007 ;; For a coding system of cateogory iso-8-1 and iso-8-2, create
1008 ;; XXX-with-esc variants.
1009 (let ((coding-category (coding-system-category coding-system)))
1010 (if (or (eq coding-category 'coding-category-iso-8-1)
1011 (eq coding-category 'coding-category-iso-8-2))
1012 (let ((esc (intern (concat (symbol-name coding-system) "-with-esc")))
1013 (doc (format "Same as %s but can handle any charsets by ISO's escape sequences." coding-system))
1014 (safe-charsets (assq 'safe-charsets properties))
1015 (mime-charset (assq 'mime-charset properties)))
1016 (if safe-charsets
1017 (setcdr safe-charsets t)
1018 (setq properties (cons (cons 'safe-charsets t) properties)))
1019 (if mime-charset
1020 (setcdr mime-charset nil))
1021 (make-coding-system esc type mnemonic doc
1022 (if (listp (car flags))
1023 (cons (append (car flags) '(t)) (cdr flags))
1024 (cons (list (car flags) t) (cdr flags)))
1025 properties))))
1026
1027 coding-system)
1028
1029 (defun define-coding-system-alias (alias coding-system)
1030 "Define ALIAS as an alias for coding system CODING-SYSTEM."
1031 (put alias 'coding-system (coding-system-spec coding-system))
1032 (nconc (coding-system-get alias 'alias-coding-systems) (list alias))
1033 (add-to-coding-system-list alias)
1034 (setq coding-system-alist (cons (list (symbol-name alias))
1035 coding-system-alist))
1036 (let ((eol-type (coding-system-eol-type coding-system)))
1037 (if (vectorp eol-type)
1038 (put alias 'eol-type (make-subsidiary-coding-system alias))
1039 (put alias 'eol-type eol-type))))
1040
1041 (defun set-buffer-file-coding-system (coding-system &optional force)
1042 "Set the file coding-system of the current buffer to CODING-SYSTEM.
1043 This means that when you save the buffer, it will be converted
1044 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
1045 use \\[list-coding-systems].
1046
1047 If the buffer's previous file coding-system value specifies end-of-line
1048 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
1049 merged with the already-specified end-of-line conversion.
1050
1051 If the buffer's previous file coding-system value specifies text
1052 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
1053 merged with the already-specified text conversion.
1054
1055 However, if the optional prefix argument FORCE is non-nil, then
1056 CODING-SYSTEM is used exactly as specified.
1057
1058 This marks the buffer modified so that the succeeding \\[save-buffer]
1059 surely saves the buffer with CODING-SYSTEM. From a program, if you
1060 don't want to mark the buffer modified, just set the variable
1061 `buffer-file-coding-system' directly."
1062 (interactive "zCoding system for visited file (default, nil): \nP")
1063 (check-coding-system coding-system)
1064 (if (and coding-system buffer-file-coding-system (null force))
1065 (let ((base (coding-system-base buffer-file-coding-system))
1066 (eol (coding-system-eol-type buffer-file-coding-system)))
1067 ;; If CODING-SYSTEM doesn't specify text conversion, merge
1068 ;; with that of buffer-file-coding-system.
1069 (if (eq (coding-system-base coding-system) 'undecided)
1070 (setq coding-system (coding-system-change-text-conversion
1071 coding-system base)))
1072 ;; If CODING-SYSTEM doesn't specify eol conversion, merge with
1073 ;; that of buffer-file-coding-system.
1074 (if (and (vectorp (coding-system-eol-type coding-system))
1075 (numberp eol) (>= eol 0) (<= eol 2))
1076 (setq coding-system (coding-system-change-eol-conversion
1077 coding-system eol)))))
1078 (setq buffer-file-coding-system coding-system)
1079 (set-buffer-modified-p t)
1080 (force-mode-line-update))
1081
1082 (defvar default-terminal-coding-system nil
1083 "Default value for the terminal coding system.
1084 This is normally set according to the selected language environment.
1085 See also the command `set-terminal-coding-system'.")
1086
1087 (defun set-terminal-coding-system (coding-system)
1088 "Set coding system of your terminal to CODING-SYSTEM.
1089 All text output to the terminal will be encoded
1090 with the specified coding system.
1091 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1092 The default is determined by the selected language environment
1093 or by the previous use of this command."
1094 (interactive
1095 (list (let ((default (if (and (not (terminal-coding-system))
1096 default-terminal-coding-system)
1097 default-terminal-coding-system)))
1098 (read-coding-system
1099 (format "Coding system for terminal display (default, %s): "
1100 default)
1101 default))))
1102 (if (and (not coding-system)
1103 (not (terminal-coding-system)))
1104 (setq coding-system default-terminal-coding-system))
1105 (if coding-system
1106 (setq default-terminal-coding-system coding-system))
1107 (set-terminal-coding-system-internal coding-system)
1108 (redraw-frame (selected-frame)))
1109
1110 (defvar default-keyboard-coding-system nil
1111 "Default value of the keyboard coding system.
1112 This is normally set according to the selected language environment.
1113 See also the command `set-keyboard-coding-system'.")
1114
1115 (defun set-keyboard-coding-system (coding-system)
1116 "Set coding system for keyboard input to CODING-SYSTEM.
1117 In addition, this command enables Encoded-kbd minor mode.
1118 \(If CODING-SYSTEM is nil, Encoded-kbd mode is turned off -- see
1119 `encoded-kbd-mode'.)
1120 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
1121 The default is determined by the selected language environment
1122 or by the previous use of this command."
1123 (interactive
1124 (list (let ((default (if (and (not (keyboard-coding-system))
1125 default-keyboard-coding-system)
1126 default-keyboard-coding-system)))
1127 (read-coding-system
1128 (format "Coding system for keyboard input (default, %s): "
1129 default)
1130 default))))
1131 (if (and (not coding-system)
1132 (not (keyboard-coding-system)))
1133 (setq coding-system default-keyboard-coding-system))
1134 (if coding-system
1135 (setq default-keyboard-coding-system coding-system))
1136 (set-keyboard-coding-system-internal coding-system)
1137 (encoded-kbd-mode (if coding-system 1 0)))
1138
1139 (defcustom keyboard-coding-system nil
1140 "Specify coding system for keyboard input.
1141 If you set this on a terminal which can't distinguish Meta keys from
1142 8-bit characters, you will have to use ESC to type Meta characters.
1143 See Info node `Specify Coding' and Info node `Single-Byte Character Support'.
1144
1145 Setting this variable directly does not take effect;
1146 use either M-x customize or \\[set-keyboard-coding-system]."
1147 :type '(coding-system :tag "Coding system")
1148 :link '(info-link "(emacs)Specify Coding")
1149 :link '(info-link "(emacs)Single-Byte Character Support")
1150 :set (lambda (symbol value)
1151 ;; Don't load encoded-kbd-mode unnecessarily.
1152 (if (or value (boundp 'encoded-kbd-mode))
1153 (set-keyboard-coding-system value)
1154 (set-default 'keyboard-coding-system nil))) ; must initialize
1155 :version "21.1"
1156 :group 'keyboard
1157 :group 'mule)
1158
1159 (defun set-buffer-process-coding-system (decoding encoding)
1160 "Set coding systems for the process associated with the current buffer.
1161 DECODING is the coding system to be used to decode input from the process,
1162 ENCODING is the coding system to be used to encode output to the process.
1163
1164 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
1165 (interactive
1166 "zCoding-system for output from the process: \nzCoding-system for input to the process: ")
1167 (let ((proc (get-buffer-process (current-buffer))))
1168 (if (null proc)
1169 (error "No process")
1170 (check-coding-system decoding)
1171 (check-coding-system encoding)
1172 (set-process-coding-system proc decoding encoding)))
1173 (force-mode-line-update))
1174
1175 (defalias 'set-clipboard-coding-system 'set-selection-coding-system)
1176
1177 (defun set-selection-coding-system (coding-system)
1178 "Make CODING-SYSTEM used for communicating with other X clients .
1179 When sending or receiving text via cut_buffer, selection, and clipboard,
1180 the text is encoded or decoded by CODING-SYSTEM."
1181 (interactive "zCoding system for X selection: ")
1182 (check-coding-system coding-system)
1183 (setq selection-coding-system coding-system))
1184
1185 ;; Coding system lastly specified by the command
1186 ;; set-next-selection-coding-system.
1187 (defvar last-next-selection-coding-system nil)
1188
1189 (defun set-next-selection-coding-system (coding-system)
1190 "Make CODING-SYSTEM used for the next communication with other X clients.
1191 This setting is effective for the next communication only."
1192 (interactive
1193 (list (read-coding-system
1194 (if last-next-selection-coding-system
1195 (format "Coding system for the next X selection (default, %S): "
1196 last-next-selection-coding-system)
1197 "Coding system for the next X selection: ")
1198 last-next-selection-coding-system)))
1199 (if coding-system
1200 (setq last-next-selection-coding-system coding-system)
1201 (setq coding-system last-next-selection-coding-system))
1202 (check-coding-system coding-system)
1203
1204 (setq next-selection-coding-system coding-system))
1205
1206 (defun set-coding-priority (arg)
1207 "Set priority of coding categories according to ARG.
1208 ARG is a list of coding categories ordered by priority."
1209 (let ((l arg)
1210 (current-list (copy-sequence coding-category-list)))
1211 ;; Check the validity of ARG while deleting coding categories in
1212 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
1213 ;; contains all coding categories.
1214 (while l
1215 (if (or (null (get (car l) 'coding-category-index))
1216 (null (memq (car l) current-list)))
1217 (error "Invalid or duplicated element in argument: %s" arg))
1218 (setq current-list (delq (car l) current-list))
1219 (setq l (cdr l)))
1220 ;; Update `coding-category-list' and return it.
1221 (setq coding-category-list (append arg current-list))
1222 (set-coding-priority-internal)))
1223
1224 ;;; FILE I/O
1225
1226 (defcustom auto-coding-alist
1227 '(("\\.\\(arc\\|zip\\|lzh\\|zoo\\|jar\\|tar\\|tgz\\)\\'" . no-conversion)
1228 ("\\.\\(ARC\\|ZIP\\|LZH\\|ZOO\\|JAR\\|TAR\\|TGZ\\)\\'" . no-conversion))
1229 "Alist of filename patterns vs corresponding coding systems.
1230 Each element looks like (REGEXP . CODING-SYSTEM).
1231 A file whose name matches REGEXP is decoded by CODING-SYSTEM on reading.
1232
1233 The settings in this alist take priority over `coding:' tags
1234 in the file (see the function `set-auto-coding')
1235 and the contents of `file-coding-system-alist'."
1236 :group 'files
1237 :group 'mule
1238 :type '(repeat (cons (regexp :tag "File name regexp")
1239 (symbol :tag "Coding system"))))
1240
1241 (defcustom auto-coding-regexp-alist
1242 '(("^BABYL OPTIONS:[ \t]*-\\*-[ \t]*rmail[ \t]*-\\*-" . no-conversion))
1243 "Alist of patterns vs corresponding coding systems.
1244 Each element looks like (REGEXP . CODING-SYSTEM).
1245 A file whose first bytes match REGEXP is decoded by CODING-SYSTEM on reading.
1246
1247 The settings in this alist take priority over `coding:' tags
1248 in the file (see the function `set-auto-coding')
1249 and the contents of `file-coding-system-alist'."
1250 :group 'files
1251 :group 'mule
1252 :type '(repeat (cons (regexp :tag "Regexp")
1253 (symbol :tag "Coding system"))))
1254
1255 (defvar set-auto-coding-for-load nil
1256 "Non-nil means look for `load-coding' property instead of `coding'.
1257 This is used for loading and byte-compiling Emacs Lisp files.")
1258
1259 (defun auto-coding-alist-lookup (filename)
1260 "Return the coding system specified by `auto-coding-alist' for FILENAME."
1261 (let ((alist auto-coding-alist)
1262 (case-fold-search (memq system-type '(vax-vms windows-nt ms-dos)))
1263 coding-system)
1264 (while (and alist (not coding-system))
1265 (if (string-match (car (car alist)) filename)
1266 (setq coding-system (cdr (car alist)))
1267 (setq alist (cdr alist))))
1268 coding-system))
1269
1270
1271 (defun auto-coding-from-file-contents (size)
1272 "Determine a coding system from the contents of the current buffer.
1273 The current buffer contains SIZE bytes starting at point.
1274 Value is either a coding system or nil."
1275 (save-excursion
1276 (let ((alist auto-coding-regexp-alist)
1277 coding-system)
1278 (while (and alist (not coding-system))
1279 (let ((regexp (car (car alist))))
1280 (when (re-search-forward regexp (+ (point) size) t)
1281 (setq coding-system (cdr (car alist)))))
1282 (setq alist (cdr alist)))
1283 coding-system)))
1284
1285
1286 (defun set-auto-coding (filename size)
1287 "Return coding system for a file FILENAME of which SIZE bytes follow point.
1288 These bytes should include at least the first 1k of the file
1289 and the last 3k of the file, but the middle may be omitted.
1290
1291 It checks FILENAME against the variable `auto-coding-alist'. If
1292 FILENAME doesn't match any entries in the variable, it checks the
1293 contents of the current buffer following point against
1294 `auto-coding-regexp-alist'. If no match is found, it checks for a
1295 `coding:' tag in the first one or two lines following point. If no
1296 `coding:' tag is found, it checks for local variables list in the last
1297 3K bytes out of the SIZE bytes.
1298
1299 The return value is the specified coding system,
1300 or nil if nothing specified.
1301
1302 The variable `set-auto-coding-function' (which see) is set to this
1303 function by default."
1304 (or (auto-coding-alist-lookup filename)
1305 (auto-coding-from-file-contents size)
1306 (let* ((case-fold-search t)
1307 (head-start (point))
1308 (head-end (+ head-start (min size 1024)))
1309 (tail-start (+ head-start (max (- size 3072) 0)))
1310 (tail-end (+ head-start size))
1311 coding-system head-found tail-found pos)
1312 ;; Try a short cut by searching for the string "coding:"
1313 ;; and for "unibyte:" at the head and tail of SIZE bytes.
1314 (setq head-found (or (search-forward "coding:" head-end t)
1315 (search-forward "unibyte:" head-end t)))
1316 (if (and head-found (> head-found tail-start))
1317 ;; Head and tail are overlapped.
1318 (setq tail-found head-found)
1319 (goto-char tail-start)
1320 (setq tail-found (or (search-forward "coding:" tail-end t)
1321 (search-forward "unibyte:" tail-end t))))
1322
1323 ;; At first check the head.
1324 (when head-found
1325 (goto-char head-start)
1326 (setq pos (re-search-forward "[\n\r]" head-end t))
1327 (if (and pos
1328 (= (char-after head-start) ?#)
1329 (= (char-after (1+ head-start)) ?!))
1330 ;; If the file begins with "#!" (exec interpreter magic),
1331 ;; look for coding frobs in the first two lines. You cannot
1332 ;; necessarily put them in the first line of such a file
1333 ;; without screwing up the interpreter invocation.
1334 (setq pos (search-forward "\n" head-end t)))
1335 (if pos (setq head-end pos))
1336 (when (< head-found head-end)
1337 (goto-char head-start)
1338 (when (and set-auto-coding-for-load
1339 (re-search-forward
1340 "-\\*-\\(.*;\\)?[ \t]*unibyte:[ \t]*\\([^ ;]+\\)"
1341 head-end t))
1342 (setq coding-system 'raw-text))
1343 (when (and (not coding-system)
1344 (re-search-forward
1345 "-\\*-\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)"
1346 head-end t))
1347 (setq coding-system (intern (match-string 2)))
1348 (or (coding-system-p coding-system)
1349 (setq coding-system nil)))))
1350
1351 ;; If no coding: tag in the head, check the tail.
1352 (when (and tail-found (not coding-system))
1353 (goto-char tail-start)
1354 (search-forward "\n\^L" nil t)
1355 (if (re-search-forward
1356 "^\\(.*\\)[ \t]*Local Variables:[ \t]*\\(.*\\)$" tail-end t)
1357 ;; The prefix is what comes before "local variables:" in its
1358 ;; line. The suffix is what comes after "local variables:"
1359 ;; in its line.
1360 (let* ((prefix (regexp-quote (match-string 1)))
1361 (suffix (regexp-quote (match-string 2)))
1362 (re-coding
1363 (concat
1364 "^" prefix
1365 "[ \t]*coding[ \t]*:[ \t]*\\([^ \t]+\\)[ \t]*"
1366 suffix "$"))
1367 (re-unibyte
1368 (concat
1369 "^" prefix
1370 "[ \t]*unibyte[ \t]*:[ \t]*\\([^ \t]+\\)[ \t]*"
1371 suffix "$"))
1372 (re-end
1373 (concat "^" prefix "[ \t]*end *:[ \t]*" suffix "$"))
1374 (pos (point)))
1375 (re-search-forward re-end tail-end 'move)
1376 (setq tail-end (point))
1377 (goto-char pos)
1378 (when (and set-auto-coding-for-load
1379 (re-search-forward re-unibyte tail-end t))
1380 (setq coding-system 'raw-text))
1381 (when (and (not coding-system)
1382 (re-search-forward re-coding tail-end t))
1383 (setq coding-system (intern (match-string 1)))
1384 (or (coding-system-p coding-system)
1385 (setq coding-system nil))))))
1386 coding-system)))
1387
1388 (setq set-auto-coding-function 'set-auto-coding)
1389
1390 (defun after-insert-file-set-buffer-file-coding-system (inserted)
1391 "Set `buffer-file-coding-system' of current buffer after text is inserted."
1392 (if last-coding-system-used
1393 (let ((coding-system
1394 (find-new-buffer-file-coding-system last-coding-system-used))
1395 (modified-p (buffer-modified-p)))
1396 (when coding-system
1397 (set-buffer-file-coding-system coding-system t)
1398 (if (and enable-multibyte-characters
1399 (or (eq coding-system 'no-conversion)
1400 (eq (coding-system-type coding-system) 5))
1401 ;; If buffer was unmodified and the size is the
1402 ;; same as INSERTED, we must be visiting it.
1403 (not modified-p)
1404 (= (buffer-size) inserted))
1405 ;; For coding systems no-conversion and raw-text...,
1406 ;; edit the buffer as unibyte.
1407 (let ((pos-byte (position-bytes (+ (point) inserted))))
1408 (set-buffer-multibyte nil)
1409 (setq inserted (- pos-byte (position-bytes (point))))))
1410 (set-buffer-modified-p modified-p))))
1411 inserted)
1412
1413 (add-hook 'after-insert-file-functions
1414 'after-insert-file-set-buffer-file-coding-system)
1415
1416 ;; The coding-spec and eol-type of coding-system returned is decided
1417 ;; independently in the following order.
1418 ;; 1. That of buffer-file-coding-system locally bound.
1419 ;; 2. That of CODING.
1420
1421 (defun find-new-buffer-file-coding-system (coding)
1422 "Return a coding system for a buffer when a file of CODING is inserted.
1423 The local variable `buffer-file-coding-system' of the current buffer
1424 is set to the returned value.
1425 Return nil if there's no need to set `buffer-file-coding-system'."
1426 (let (local-coding local-eol
1427 found-coding found-eol
1428 new-coding new-eol)
1429 (if (null coding)
1430 ;; Nothing found about coding.
1431 nil
1432
1433 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
1434 ;; and LOCAL-CODING.
1435 (setq local-eol (coding-system-eol-type buffer-file-coding-system))
1436 (if (null (numberp local-eol))
1437 ;; But eol-type is not yet set.
1438 (setq local-eol nil))
1439 (if (and buffer-file-coding-system
1440 (not (eq (coding-system-type buffer-file-coding-system) t)))
1441 ;; This is not `undecided'.
1442 (setq local-coding (coding-system-base buffer-file-coding-system)))
1443
1444 (if (and (local-variable-p 'buffer-file-coding-system)
1445 local-eol local-coding)
1446 ;; The current buffer has already set full coding-system, we
1447 ;; had better not change it.
1448 nil
1449
1450 (setq found-eol (coding-system-eol-type coding))
1451 (if (null (numberp found-eol))
1452 ;; But eol-type is not found.
1453 ;; If EOL conversions are inhibited, force unix eol-type.
1454 (setq found-eol (if inhibit-eol-conversion 0)))
1455 (if (eq (coding-system-type coding) t)
1456 (setq found-coding 'undecided)
1457 (setq found-coding (coding-system-base coding)))
1458
1459 (if (and (not found-eol) (eq found-coding 'undecided))
1460 ;; No valid coding information found.
1461 nil
1462
1463 ;; Some coding information (eol or text) found.
1464
1465 ;; The local setting takes precedence over the found one.
1466 (setq new-coding (if (local-variable-p 'buffer-file-coding-system)
1467 (or local-coding found-coding)
1468 (or found-coding local-coding)))
1469 (setq new-eol (if (local-variable-p 'buffer-file-coding-system)
1470 (or local-eol found-eol)
1471 (or found-eol local-eol)))
1472
1473 (let ((eol-type (coding-system-eol-type new-coding)))
1474 (if (and (numberp new-eol) (vectorp eol-type))
1475 (aref eol-type new-eol)
1476 new-coding)))))))
1477
1478 (defun modify-coding-system-alist (target-type regexp coding-system)
1479 "Modify one of look up tables for finding a coding system on I/O operation.
1480 There are three of such tables, `file-coding-system-alist',
1481 `process-coding-system-alist', and `network-coding-system-alist'.
1482
1483 TARGET-TYPE specifies which of them to modify.
1484 If it is `file', it affects `file-coding-system-alist' (which see).
1485 If it is `process', it affects `process-coding-system-alist' (which see).
1486 If it is `network', it affects `network-coding-system-alist' (which see).
1487
1488 REGEXP is a regular expression matching a target of I/O operation.
1489 The target is a file name if TARGET-TYPE is `file', a program name if
1490 TARGET-TYPE is `process', or a network service name or a port number
1491 to connect to if TARGET-TYPE is `network'.
1492
1493 CODING-SYSTEM is a coding system to perform code conversion on the I/O
1494 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
1495 for decoding and encoding respectively,
1496 or a function symbol which, when called, returns such a cons cell."
1497 (or (memq target-type '(file process network))
1498 (error "Invalid target type: %s" target-type))
1499 (or (stringp regexp)
1500 (and (eq target-type 'network) (integerp regexp))
1501 (error "Invalid regular expression: %s" regexp))
1502 (if (symbolp coding-system)
1503 (if (not (fboundp coding-system))
1504 (progn
1505 (check-coding-system coding-system)
1506 (setq coding-system (cons coding-system coding-system))))
1507 (check-coding-system (car coding-system))
1508 (check-coding-system (cdr coding-system)))
1509 (cond ((eq target-type 'file)
1510 (let ((slot (assoc regexp file-coding-system-alist)))
1511 (if slot
1512 (setcdr slot coding-system)
1513 (setq file-coding-system-alist
1514 (cons (cons regexp coding-system)
1515 file-coding-system-alist)))))
1516 ((eq target-type 'process)
1517 (let ((slot (assoc regexp process-coding-system-alist)))
1518 (if slot
1519 (setcdr slot coding-system)
1520 (setq process-coding-system-alist
1521 (cons (cons regexp coding-system)
1522 process-coding-system-alist)))))
1523 (t
1524 (let ((slot (assoc regexp network-coding-system-alist)))
1525 (if slot
1526 (setcdr slot coding-system)
1527 (setq network-coding-system-alist
1528 (cons (cons regexp coding-system)
1529 network-coding-system-alist)))))))
1530
1531 (defun make-translation-table (&rest args)
1532 "Make a translation table from arguments.
1533 A translation table is a char table intended for for character
1534 translation in CCL programs.
1535
1536 Each argument is a list of elemnts of the form (FROM . TO), where FROM
1537 is a character to be translated to TO.
1538
1539 FROM can be a generic character (see `make-char'). In this case, TO is
1540 a generic character containing the same number of characters, or a
1541 ordinary character. If FROM and TO are both generic characters, all
1542 characters belonging to FROM are translated to characters belonging to TO
1543 without changing their position code(s).
1544
1545 The arguments and forms in each argument are processed in the given
1546 order, and if a previous form already translates TO to some other
1547 character, say TO-ALT, FROM is also translated to TO-ALT."
1548 (let ((table (make-char-table 'translation-table))
1549 revlist)
1550 (while args
1551 (let ((elts (car args)))
1552 (while elts
1553 (let* ((from (car (car elts)))
1554 (from-i 0) ; degree of freedom of FROM
1555 (from-rev (nreverse (split-char from)))
1556 (to (cdr (car elts)))
1557 (to-i 0) ; degree of freedom of TO
1558 (to-rev (nreverse (split-char to))))
1559 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
1560 (while (eq (car from-rev) 0)
1561 (setq from-i (1+ from-i) from-rev (cdr from-rev)))
1562 (while (eq (car to-rev) 0)
1563 (setq to-i (1+ to-i) to-rev (cdr to-rev)))
1564 (if (and (/= from-i to-i) (/= to-i 0))
1565 (error "Invalid character pair (%d . %d)" from to))
1566 ;; If we have already translated TO to TO-ALT, FROM should
1567 ;; also be translated to TO-ALT. But, this is only if TO
1568 ;; is a generic character or TO-ALT is not a generic
1569 ;; character.
1570 (let ((to-alt (aref table to)))
1571 (if (and to-alt
1572 (or (> to-i 0) (not (generic-char-p to-alt))))
1573 (setq to to-alt)))
1574 (if (> from-i 0)
1575 (set-char-table-default table from to)
1576 (aset table from to))
1577 ;; If we have already translated some chars to FROM, they
1578 ;; should also be translated to TO.
1579 (let ((l (assq from revlist)))
1580 (if l
1581 (let ((ch (car l)))
1582 (setcar l to)
1583 (setq l (cdr l))
1584 (while l
1585 (aset table ch to)
1586 (setq l (cdr l)) ))))
1587 ;; Now update REVLIST.
1588 (let ((l (assq to revlist)))
1589 (if l
1590 (setcdr l (cons from (cdr l)))
1591 (setq revlist (cons (list to from) revlist)))))
1592 (setq elts (cdr elts))))
1593 (setq args (cdr args)))
1594 ;; Return TABLE just created.
1595 table))
1596
1597 (defun make-translation-table-from-vector (vec)
1598 "Make translation table from decoding vector VEC.
1599 VEC is an array of 256 elements to map unibyte codes to multibyte characters.
1600 See also the variable `nonascii-translation-table'."
1601 (let ((table (make-char-table 'translation-table))
1602 (rev-table (make-char-table 'translation-table))
1603 (i 0)
1604 ch)
1605 (while (< i 256)
1606 (setq ch (aref vec i))
1607 (aset table i ch)
1608 (if (>= ch 256)
1609 (aset rev-table ch i))
1610 (setq i (1+ i)))
1611 (set-char-table-extra-slot table 0 rev-table)
1612 table))
1613
1614 (defun define-translation-table (symbol &rest args)
1615 "Define SYMBOL as the name of translation table made by ARGS.
1616 This sets up information so that the table can be used for
1617 translations in a CCL program.
1618
1619 If the first element of ARGS is a char-table whose purpose is
1620 `translation-table', just define SYMBOL to name it. (Note that this
1621 function does not bind SYMBOL.)
1622
1623 Any other ARGS should be suitable as arguments of the function
1624 `make-translation-table' (which see).
1625
1626 This function sets properties `translation-table' and
1627 `translation-table-id' of SYMBOL to the created table itself and the
1628 identification number of the table respectively. It also registers
1629 the table in `translation-table-vector'."
1630 (let ((table (if (and (char-table-p (car args))
1631 (eq (char-table-subtype (car args))
1632 'translation-table))
1633 (car args)
1634 (apply 'make-translation-table args)))
1635 (len (length translation-table-vector))
1636 (id 0)
1637 (done nil))
1638 (put symbol 'translation-table table)
1639 (while (not done)
1640 (if (>= id len)
1641 (setq translation-table-vector
1642 (vconcat translation-table-vector (make-vector len nil))))
1643 (let ((slot (aref translation-table-vector id)))
1644 (if (or (not slot)
1645 (eq (car slot) symbol))
1646 (progn
1647 (aset translation-table-vector id (cons symbol table))
1648 (setq done t))
1649 (setq id (1+ id)))))
1650 (put symbol 'translation-table-id id)
1651 id))
1652
1653 (put 'with-category-table 'lisp-indent-function 1)
1654
1655 (defmacro with-category-table (category-table &rest body)
1656 `(let ((current-category-table (category-table)))
1657 (set-category-table ,category-table)
1658 (unwind-protect
1659 (progn ,@body)
1660 (set-category-table current-category-table))))
1661
1662 ;;; Initialize some variables.
1663
1664 (put 'use-default-ascent 'char-table-extra-slots 0)
1665 (setq use-default-ascent (make-char-table 'use-default-ascent))
1666 (put 'ignore-relative-composition 'char-table-extra-slots 0)
1667 (setq ignore-relative-composition
1668 (make-char-table 'ignore-relative-composition))
1669
1670 ;;;
1671 (provide 'mule)
1672
1673 ;;; mule.el ends here