1 ;;; mule.el --- basic commands for mulitilingual environment
3 ;; Copyright (C) 1995 Electrotechnical Laboratory, JAPAN.
4 ;; Licensed to the Free Software Foundation.
6 ;; Keywords: mule, multilingual, character set, coding system
8 ;; This file is part of GNU Emacs.
10 ;; GNU Emacs is free software; you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation; either version 2, or (at your option)
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs; see the file COPYING. If not, write to the
22 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 ;; Boston, MA 02111-1307, USA.
27 (defconst mule-version
"3.0 (MOMIJINOGA)" "\
28 Version number and name of this version of MULE (multilingual environment).")
30 (defconst mule-version-date
"1998.1.1" "\
31 Distribution date of this version of MULE (multilingual environment).")
33 (defun load-with-code-conversion (fullname file
&optional noerror nomessage
)
34 "Execute a file of Lisp code named FILE whose absolute name is FULLNAME.
35 The file contents are decoded before evaluation if necessary.
36 If optional second arg NOERROR is non-nil,
37 report no error if FILE doesn't exist.
38 Print messages at start and end of loading unless
39 optional third arg NOMESSAGE is non-nil.
40 Return t if file exists."
41 (if (null (file-readable-p fullname
))
43 (signal 'file-error
(list "Cannot open load file" file
)))
44 ;; Read file with code conversion, and then eval.
46 ;; To avoid any autoloading, set default-major-mode to
48 (let ((default-major-mode 'fundamental-mode
))
49 ;; We can't use `generate-new-buffer' because files.el
51 (get-buffer-create (generate-new-buffer-name " *load*"))))
53 (source (save-match-data (string-match "\\.el\\'" fullname
))))
56 (message "Loading %s (source)..." file
)
57 (message "Loading %s..." file
)))
59 (setq preloaded-file-list
(cons file preloaded-file-list
)))
61 (let ((load-file-name fullname
)
62 (inhibit-file-name-operation nil
))
65 ;; This is buffer-local.
66 (setq enable-multibyte-characters t
)
67 (insert-file-contents fullname
)
68 ;; Make `kill-buffer' quiet.
69 (set-buffer-modified-p nil
))
70 ;; Have the original buffer current while we eval.
71 (eval-buffer buffer nil file
))
72 (let (kill-buffer-hook kill-buffer-query-functions
)
73 (kill-buffer buffer
)))
74 (let ((hook (assoc file after-load-alist
)))
76 (mapcar (function eval
) (cdr hook
))))
77 (unless (or nomessage noninteractive
)
79 (message "Loading %s (source)...done" file
)
80 (message "Loading %s...done" file
)))
83 ;; API (Application Program Interface) for charsets.
85 ;; Return t if OBJ is a quoted symbol
86 ;; and the symbol is the name of a standard charset.
87 (defsubst charset-quoted-standard-p
(obj)
88 (and (listp obj
) (eq (car obj
) 'quote
)
89 (symbolp (car-safe (cdr obj
)))
90 (let ((vector (get (car-safe (cdr obj
)) 'charset
)))
92 (< (aref vector
0) 160)))))
94 (defsubst charsetp
(object)
95 "T is OBJECT is a charset."
96 (and (symbolp object
) (vectorp (get object
'charset
))))
98 (defsubst charset-info
(charset)
99 "Return a vector of information of CHARSET.
100 The elements of the vector are:
101 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
102 LEADING-CODE-BASE, LEADING-CODE-EXT,
103 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
104 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
107 CHARSET-ID (integer) is the identification number of the charset.
108 DIMENSION (integer) is the number of bytes to represent a character of
110 CHARS (integer) is the number of characters in a dimension: 94 or 96.
111 BYTE (integer) is the length of multi-byte form of a character in
112 the charset: one of 1, 2, 3, and 4.
113 WIDTH (integer) is the number of columns a character in the charset
114 occupies on the screen: one of 0, 1, and 2.
115 DIRECTION (integer) is the rendering direction of characters in the
116 charset when rendering. If 0, render from right to left, else
117 render from left to right.
118 LEADING-CODE-BASE (integer) is the base leading-code for the
120 LEADING-CODE-EXT (integer) is the extended leading-code for the
121 charset. All charsets of less than 0xA0 has the value 0.
122 ISO-FINAL-CHAR (character) is the final character of the
123 corresponding ISO 2022 charset.
124 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
125 while encoding to variants of ISO 2022 coding system, one of the
126 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
127 REVERSE-CHARSET (integer) is the charset which differs only in
128 LEFT-TO-RIGHT value from the charset. If there's no such a
129 charset, the value is -1.
130 SHORT-NAME (string) is the short name to refer to the charset.
131 LONG-NAME (string) is the long name to refer to the charset
132 DESCRIPTION (string) is the description string of the charset.
133 PLIST (property list) may contain any type of information a user
134 want to put and get by functions `put-charset-property' and
135 `get-charset-property' respectively."
136 (get charset
'charset
))
138 (defmacro charset-id
(charset)
139 "Return charset identification number of CHARSET."
140 (if (charset-quoted-standard-p charset
)
141 (aref (charset-info (nth 1 charset
)) 0)
142 `(aref (charset-info ,charset
) 0)))
144 (defmacro charset-bytes
(charset)
145 "Return bytes of CHARSET.
146 See the function `charset-info' for more detail."
147 (if (charset-quoted-standard-p charset
)
148 (aref (charset-info (nth 1 charset
)) 1)
149 `(aref (charset-info ,charset
) 1)))
151 (defmacro charset-dimension
(charset)
152 "Return dimension of CHARSET.
153 See the function `charset-info' for more detail."
154 (if (charset-quoted-standard-p charset
)
155 (aref (charset-info (nth 1 charset
)) 2)
156 `(aref (charset-info ,charset
) 2)))
158 (defmacro charset-chars
(charset)
159 "Return character numbers contained in a dimension of CHARSET.
160 See the function `charset-info' for more detail."
161 (if (charset-quoted-standard-p charset
)
162 (aref (charset-info (nth 1 charset
)) 3)
163 `(aref (charset-info ,charset
) 3)))
165 (defmacro charset-width
(charset)
166 "Return width (how many column occupied on a screen) of CHARSET.
167 See the function `charset-info' for more detail."
168 (if (charset-quoted-standard-p charset
)
169 (aref (charset-info (nth 1 charset
)) 4)
170 `(aref (charset-info ,charset
) 4)))
172 (defmacro charset-direction
(charset)
173 "Return direction of CHARSET.
174 See the function `charset-info' for more detail."
175 (if (charset-quoted-standard-p charset
)
176 (aref (charset-info (nth 1 charset
)) 5)
177 `(aref (charset-info ,charset
) 5)))
179 (defmacro charset-iso-final-char
(charset)
180 "Return final char of CHARSET.
181 See the function `charset-info' for more detail."
182 (if (charset-quoted-standard-p charset
)
183 (aref (charset-info (nth 1 charset
)) 8)
184 `(aref (charset-info ,charset
) 8)))
186 (defmacro charset-iso-graphic-plane
(charset)
187 "Return graphic plane of CHARSET.
188 See the function `charset-info' for more detail."
189 (if (charset-quoted-standard-p charset
)
190 (aref (charset-info (nth 1 charset
)) 9)
191 `(aref (charset-info ,charset
) 9)))
193 (defmacro charset-reverse-charset
(charset)
194 "Return reverse charset of CHARSET.
195 See the function `charset-info' for more detail."
196 (if (charset-quoted-standard-p charset
)
197 (aref (charset-info (nth 1 charset
)) 10)
198 `(aref (charset-info ,charset
) 10)))
200 (defmacro charset-short-name
(charset)
201 "Return short name of CHARSET.
202 See the function `charset-info' for more detail."
203 (if (charset-quoted-standard-p charset
)
204 (aref (charset-info (nth 1 charset
)) 11)
205 `(aref (charset-info ,charset
) 11)))
207 (defmacro charset-long-name
(charset)
208 "Return long name of CHARSET.
209 See the function `charset-info' for more detail."
210 (if (charset-quoted-standard-p charset
)
211 (aref (charset-info (nth 1 charset
)) 12)
212 `(aref (charset-info ,charset
) 12)))
214 (defmacro charset-description
(charset)
215 "Return descriptoin of CHARSET.
216 See the function `charset-info' for more detail."
217 (if (charset-quoted-standard-p charset
)
218 (aref (charset-info (nth 1 charset
)) 13)
219 `(aref (charset-info ,charset
) 13)))
221 (defmacro charset-plist
(charset)
222 "Return list charset property of CHARSET.
223 See the function `charset-info' for more detail."
224 (if (charset-quoted-standard-p charset
)
225 `(aref ,(charset-info (nth 1 charset
)) 14)
226 `(aref (charset-info ,charset
) 14)))
228 (defun set-charset-plist (charset plist
)
229 "Set CHARSET's property list to PLIST, and retrun PLIST."
230 (aset (charset-info charset
) 14 plist
))
232 (defun make-char (charset &optional c1 c2
)
233 "Return a character of CHARSET and position-codes CODE1 and CODE2.
234 CODE1 and CODE2 are optional, but if you don't supply
235 sufficient position-codes, return a generic character which stands for
236 all characters or group of characters in the character sets.
237 A generic character can be used to index a char table (e.g. syntax-table)."
238 (make-char-internal (charset-id charset
) c1 c2
))
240 (put 'make-char
'byte-compile
243 (let ((charset (nth 1 form
)))
244 (if (charset-quoted-standard-p charset
)
245 (byte-compile-normal-call
246 (cons 'make-char-internal
247 (cons (charset-id (nth 1 charset
)) (nthcdr 2 form
))))
248 (byte-compile-normal-call
249 (cons 'make-char-internal
250 (cons (list 'charset-id charset
) (nthcdr 2 form
)))))))))
252 (defun charset-list ()
253 "Return list of charsets ever defined.
255 This function is provided for backward compatibility.
256 Now we have the variable `charset-list'."
259 (make-obsolete 'charset-list
260 "Use the variable charset-list instead.")
262 (defsubst generic-char-p
(char)
263 "Return t if and only if CHAR is a generic character.
264 See also the documentation of make-char."
265 (let ((l (split-char char
)))
266 (and (or (= (nth 1 l
) 0) (eq (nth 2 l
) 0))
267 (not (eq (car l
) 'composition
)))))
270 ;; Coding system staffs
272 ;; Coding system is a symbol that has the property `coding-system'.
274 ;; The value of the property `coding-system' is a vector of the
276 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
277 ;; We call this vector as coding-spec. See comments in src/coding.c
280 (defconst coding-spec-type-idx
0)
281 (defconst coding-spec-mnemonic-idx
1)
282 (defconst coding-spec-doc-string-idx
2)
283 (defconst coding-spec-plist-idx
3)
284 (defconst coding-spec-flags-idx
4)
286 ;; PLIST is a property list of a coding system. To share PLIST among
287 ;; alias coding systems, a coding system has PLIST in coding-spec
288 ;; instead of having it in normal property list of Lisp symbol.
289 ;; Here's a list of coding system properties currently being used.
293 ;; The value is a coding category the coding system belongs to. The
294 ;; function `make-coding-system' and `define-coding-system-alias' sets
295 ;; this value automatically.
297 ;; o alias-coding-systems
299 ;; The value is a list of coding systems of the same alias group. The
300 ;; first element is the coding system made at first, which we call as
301 ;; `base coding system'. The function `make-coding-system' and
302 ;; `define-coding-system-alias' set this value automatically.
304 ;; o post-read-conversion
306 ;; The value is a function to call after some text is inserted and
307 ;; decoded by the coding system itself and before any functions in
308 ;; `after-insert-functions' are called. The arguments to this
309 ;; function is the same as those of a function in
310 ;; `after-insert-functions', i.e. LENGTH of a text while putting point
311 ;; at the head of the text to be decoded
313 ;; o pre-write-conversion
315 ;; The value is a function to call after all functions in
316 ;; `write-region-annotate-functions' and `buffer-file-format' are
317 ;; called, and before the text is encoded by the coding system itself.
318 ;; The arguments to this function is the same as those of a function
319 ;; in `write-region-annotate-functions', i.e. FROM and TO specifying
322 ;; o character-unification-table-for-decode
324 ;; The value is a unification table to be applied on decoding. See
325 ;; the function `make-unification-table' for the format of unification
328 ;; o character-unification-table-for-encode
330 ;; The value is a unification table to be applied on encoding.
334 ;; The value is a list of charsets safely supported by the coding
335 ;; system. The value t means that all charsets Emacs handles are
336 ;; supported. Even if some charset is not in this list, it doesn't
337 ;; mean that the charset can't be encoded in the coding system,
338 ;; instead, it just means that some other receiver of a text encoded
339 ;; in the coding system won't be able to handle that charset.
342 ;; Return coding-spec of CODING-SYSTEM
343 (defsubst coding-system-spec
(coding-system)
344 (get (check-coding-system coding-system
) 'coding-system
))
346 (defun coding-system-type (coding-system)
347 "Return the coding type of CODING-SYSTEM.
348 A coding type is an integer value indicating the encoding method
349 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
350 (aref (coding-system-spec coding-system
) coding-spec-type-idx
))
352 (defun coding-system-mnemonic (coding-system)
353 "Return the mnemonic character of CODING-SYSTEM.
354 A mnemonic character of a coding system is used in mode line
355 to indicate the coding system."
356 (or (aref (coding-system-spec coding-system
) coding-spec-mnemonic-idx
)
359 (defun coding-system-doc-string (coding-system)
360 "Return the documentation string for CODING-SYSTEM."
361 (aref (coding-system-spec coding-system
) coding-spec-doc-string-idx
))
363 (defun coding-system-plist (coding-system)
364 "Return the property list of CODING-SYSTEM."
365 (aref (coding-system-spec coding-system
) coding-spec-plist-idx
))
367 (defun coding-system-flags (coding-system)
368 "Return `flags' of CODING-SYSTEM.
369 A `flags' of a coding system is a vector of length 32 indicating detailed
370 information of a coding system. See the function `make-coding-system'
372 (aref (coding-system-spec coding-system
) coding-spec-flags-idx
))
374 (defun coding-system-get (coding-system prop
)
375 "Extract a value from CODING-SYSTEM's property list for property PROP."
376 (plist-get (coding-system-plist coding-system
) prop
))
378 (defun coding-system-put (coding-system prop val
)
379 "Change value in CODING-SYSTEM's property list PROP to VAL."
380 (let ((plist (coding-system-plist coding-system
)))
382 (plist-put plist prop val
)
383 (aset (coding-system-spec coding-system
) coding-spec-plist-idx
386 (defun coding-system-category (coding-system)
387 "Return the coding category of CODING-SYSTEM."
388 (coding-system-get coding-system
'coding-category
))
390 (defun coding-system-base (coding-system)
391 "Return the base coding system of CODING-SYSTEM.
392 A base coding system is what made by `make-coding-system',
393 not what made by `define-coding-system-alias'."
394 (car (coding-system-get coding-system
'alias-coding-systems
)))
396 (defalias 'coding-system-parent
'coding-system-base
)
397 (make-obsolete 'coding-system-parent
'coding-system-base
)
399 ;; Coding system also has a property `eol-type'.
401 ;; This property indicates how the coding system handles end-of-line
402 ;; format. The value is integer 0, 1, 2, or a vector of three coding
403 ;; systems. Each integer value 0, 1, and 2 indicates the format of
404 ;; end-of-line LF, CRLF, and CR respectively. A vector value
405 ;; indicates that the format of end-of-line should be detected
406 ;; automatically. Nth element of the vector is the subsidiary coding
407 ;; system whose `eol-type' property is N.
409 (defun coding-system-eol-type (coding-system)
410 "Return eol-type of CODING-SYSTEM.
411 An eol-type is integer 0, 1, 2, or a vector of coding systems.
413 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
414 CRLF, and CR respectively.
416 A vector value indicates that a format of end-of-line should be
417 detected automatically. Nth element of the vector is the subsidiary
418 coding system whose eol-type is N."
419 (get coding-system
'eol-type
))
421 ;; Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM.
422 (defun make-subsidiary-coding-system (coding-system)
423 (let ((coding-spec (coding-system-spec coding-system
))
424 (subsidiaries (vector (intern (format "%s-unix" coding-system
))
425 (intern (format "%s-dos" coding-system
))
426 (intern (format "%s-mac" coding-system
))))
430 (put (aref subsidiaries i
) 'coding-system coding-spec
)
431 (put (aref subsidiaries i
) 'eol-type i
)
432 (setq coding-system-list
433 (cons (aref subsidiaries i
) coding-system-list
))
434 (setq coding-system-alist
435 (cons (list (symbol-name (aref subsidiaries i
)))
436 coding-system-alist
))
440 (defun make-coding-system (coding-system type mnemonic doc-string
441 &optional flags safe-charsets
)
442 "Define a new CODING-SYSTEM (symbol).
443 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
444 and CHARSETS (optional) which construct a coding-spec of CODING-SYSTEM
445 in the following format:
446 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
447 TYPE is an integer value indicating the type of coding-system as follows:
448 0: Emacs internal format,
449 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PC,
450 2: ISO-2022 including many variants,
451 3: Big5 used mainly on Chinese PC,
452 4: private, CCL programs provide encoding/decoding algorithm,
453 5: Raw-text, which means that text contains random 8-bit codes.
455 MNEMONIC is a character to be displayed on mode line for the coding-system.
457 DOC-STRING is a documentation string for the coding-system.
459 PLIST is the propert list for CODING-SYSTEM. This function sets
460 properties coding-category, alias-coding-systems, safe-charsets. The
461 first two are set automatically. The last one is set to the argument
462 SAFE-CHARSETS. SAFE-CHARSETS is a list of character sets encoded
463 safely in CODING-SYSTEM, or t which means all character sets are safe.
465 FLAGS specifies more precise information of each TYPE.
467 If TYPE is 2 (ISO-2022), FLAGS should be a list of:
468 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
469 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
470 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
471 SAFE, ACCEPT-LATIN-EXTRA-CODE.
472 CHARSETn are character sets initially designated to Gn graphic registers.
473 If CHARSETn is nil, Gn is never used.
474 If CHARSETn is t, Gn can be used but nothing designated initially.
475 If CHARSETn is a list of character sets, those character sets are
476 designated to Gn on output, but nothing designated to Gn initially.
477 SHORT-FORM non-nil means use short designation sequence on output.
478 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
479 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
481 SEVEN non-nil means use 7-bit code only on output.
482 LOCKING-SHIFT non-nil means use locking-shift.
483 SINGLE-SHIFT non-nil means use single-shift.
484 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
485 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
486 NO-ISO6429 non-nil means not use ISO6429's direction specification.
487 INIT-BOL non-nil means any designation state is assumed to be reset
488 to initial at each beginning of line on output.
489 DESIGNATION-BOL non-nil means designation sequences should be placed
490 at beginning of line on output.
491 SAFE non-nil means convert unsafe characters to `?' on output.
492 Unsafe characters are what not specified in SAFE-CHARSET.
493 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
494 a code specified in `latin-extra-code-table' (which see) as a valid
495 code of the coding system.
497 If TYPE is 4 (private), FLAGS should be a cons of CCL programs,
498 for decoding and encoding. See the documentation of CCL for more detail."
500 (if (memq coding-system coding-system-list
)
501 (error "Coding system %s already exists"))
503 ;; Set a value of `coding-system' property.
504 (let ((coding-spec (make-vector 5 nil
))
505 (no-initial-designation nil
)
507 (if (or (not (integerp type
)) (< type
0) (> type
5))
508 (error "TYPE argument must be 0..5"))
509 (if (or (not (integerp mnemonic
)) (<= mnemonic ?
) (> mnemonic
127))
510 (error "MNEMONIC arguemnt must be an ASCII printable character."))
511 (aset coding-spec coding-spec-type-idx type
)
512 (aset coding-spec coding-spec-mnemonic-idx mnemonic
)
513 (aset coding-spec coding-spec-doc-string-idx
514 (if (stringp doc-string
) doc-string
""))
516 (setq coding-category
'coding-category-emacs-mule
))
518 (setq coding-category
'coding-category-sjis
))
519 ((= type
2) ; ISO2022
521 (vec (make-vector 32 nil
))
522 (g1-designation nil
))
523 (setq no-initial-designation t
)
525 (let ((charset (car flags
)))
526 (if (and no-initial-designation
528 (or (charsetp charset
)
530 (charsetp (car charset
)))))
531 (setq no-initial-designation nil
))
532 (if (charsetp charset
)
533 (if (= i
1) (setq g1-designation charset
))
538 (setq elt
(car tail
))
539 (or (not elt
) (eq elt t
) (charsetp elt
)
540 (error "Invalid charset: %s" elt
))
541 (setq tail
(cdr tail
)))
542 (setq g1-designation
(car charset
)))
543 (if (and charset
(not (eq charset t
)))
544 (error "Invalid charset: %s" charset
))))
545 (aset vec i charset
))
546 (setq flags
(cdr flags
) i
(1+ i
)))
547 (while (and (< i
32) flags
)
548 (aset vec i
(car flags
))
549 (setq flags
(cdr flags
) i
(1+ i
)))
550 (aset coding-spec
4 vec
)
551 (setq coding-category
552 (if (aref vec
8) ; Use locking-shift.
553 (or (and (aref vec
7) 'coding-category-iso-7-else
)
554 'coding-category-iso-8-else
)
555 (if (aref vec
7) ; 7-bit only.
556 (if (aref vec
9) ; Use single-shift.
557 'coding-category-iso-7-else
558 'coding-category-iso-7
)
559 (if no-initial-designation
560 'coding-category-iso-8-else
561 (if (and (charsetp g1-designation
)
562 (= (charset-dimension g1-designation
) 2))
563 'coding-category-iso-8-2
564 'coding-category-iso-8-1
)))))))
566 (setq coding-category
'coding-category-big5
))
567 ((= type
4) ; private
568 (setq coding-category
'coding-category-binary
)
569 (if (and (consp flags
)
570 (vectorp (car flags
))
571 (vectorp (cdr flags
)))
572 (aset coding-spec
4 flags
)
573 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))
575 (setq coding-category
'coding-category-raw-text
)))
577 (let ((plist (list 'coding-category coding-category
578 'alias-coding-systems
(list coding-system
)
579 'safe-charsets safe-charsets
)))
580 (if no-initial-designation
581 (setq plist
(cons 'no-initial-designation
582 (cons no-initial-designation plist
))))
583 (aset coding-spec coding-spec-plist-idx plist
))
584 (put coding-system
'coding-system coding-spec
)
585 (put coding-category
'coding-systems
586 (cons coding-system
(get coding-category
'coding-systems
))))
588 ;; Next, set a value of `eol-type' property. The value is a vector
589 ;; of subsidiary coding systems, each corresponds to a coding system
590 ;; for the detected end-of-line format.
591 (put coding-system
'eol-type
592 (if (or (<= type
3) (= type
5))
593 (make-subsidiary-coding-system coding-system
)
596 ;; At last, register CODING-SYSTEM in `coding-system-list' and
597 ;; `coding-system-alist'.
598 (setq coding-system-list
(cons coding-system coding-system-list
))
599 (setq coding-system-alist
(cons (list (symbol-name coding-system
))
600 coding-system-alist
)))
602 (defun define-coding-system-alias (alias coding-system
)
603 "Define ALIAS as an alias for coding system CODING-SYSTEM."
604 (put alias
'coding-system
(coding-system-spec coding-system
))
605 (nconc (coding-system-get alias
'alias-coding-systems
) (list alias
))
606 (setq coding-system-list
(cons alias coding-system-list
))
607 (setq coding-system-alist
(cons (list (symbol-name alias
))
608 coding-system-alist
))
609 (let ((eol-type (coding-system-eol-type coding-system
)))
610 (if (vectorp eol-type
)
611 (put alias
'eol-type
(make-subsidiary-coding-system alias
))
612 (put alias
'eol-type eol-type
))))
614 (defun set-buffer-file-coding-system (coding-system &optional force
)
615 "Set the file coding-system of the current buffer to CODING-SYSTEM.
616 This means that when you save the buffer, it will be converted
617 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
618 use \\[list-coding-systems].
620 If the buffer's previous file coding-system value specifies end-of-line
621 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
622 merged with the already-specified end-of-line conversion.
623 However, if the optional prefix argument FORCE is non-nil,
624 then CODING-SYSTEM is used exactly as specified."
625 (interactive "zCoding system for visited file: \nP")
626 (check-coding-system coding-system
)
628 (let ((x (coding-system-eol-type buffer-file-coding-system
))
629 (y (coding-system-eol-type coding-system
)))
630 (if (and (numberp x
) (>= x
0) (<= x
2) (vectorp y
))
631 (setq coding-system
(aref y x
)))))
632 (setq buffer-file-coding-system coding-system
)
633 (set-buffer-modified-p t
)
634 (force-mode-line-update))
636 (defvar default-terminal-coding-system nil
637 "Default value for the terminal coding system.
638 This is normally set according to the selected language environment.
639 See also the command `set-terminal-coding-system'.")
641 (defun set-terminal-coding-system (coding-system)
642 "Set coding system of your terminal to CODING-SYSTEM.
643 All text output to the terminal will be encoded
644 with the specified coding system.
645 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
646 The default is determined by the selected language environment
647 or by the previous use of this command."
649 (list (let ((default (if (and (not (terminal-coding-system))
650 default-terminal-coding-system
)
651 default-terminal-coding-system
)))
653 (format "Coding system for terminal display (default, %s): "
656 (if (and (not coding-system
)
657 (not (terminal-coding-system)))
658 (setq coding-system default-terminal-coding-system
))
660 (setq default-terminal-coding-system coding-system
))
661 (set-terminal-coding-system-internal coding-system
)
662 (redraw-frame (selected-frame)))
664 (defvar default-keyboard-coding-system nil
665 "Default value of the keyboard coding system.
666 This is normally set according to the selected language environment.
667 See also the command `set-keyboard-coding-system'.")
669 (defun set-keyboard-coding-system (coding-system)
670 "Set coding system for keyboard input to CODING-SYSTEM.
671 In addition, this command enables Encoded-kbd minor mode.
672 \(If CODING-SYSTEM is nil, Encoded-bkd mode is turned off.)
673 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
674 The default is determined by the selected language environment
675 or by the previous use of this command."
677 (list (let ((default (if (and (not (keyboard-coding-system))
678 default-keyboard-coding-system
)
679 default-keyboard-coding-system
)))
681 (format "Coding system for keyboard input (default, %s): "
684 (if (and (not coding-system
)
685 (not (keyboard-coding-system)))
686 (setq coding-system default-keyboard-coding-system
))
688 (setq default-keyboard-coding-system coding-system
))
689 (set-keyboard-coding-system-internal coding-system
)
690 (encoded-kbd-mode (if coding-system
1 0)))
692 (defun set-buffer-process-coding-system (decoding encoding
)
693 "Set coding systems for the process associated with the current buffer.
694 DECODING is the coding system to be used to decode input from the process,
695 ENCODING is the coding system to be used to encode output to the process.
697 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
699 "zCoding-system for process input: \nzCoding-system for process output: ")
700 (let ((proc (get-buffer-process (current-buffer))))
703 (check-coding-system decoding
)
704 (check-coding-system encoding
)
705 (set-process-coding-system proc decoding encoding
)))
706 (force-mode-line-update))
708 (defun set-coding-priority (arg)
709 "Set priority of coding-category according to LIST.
710 LIST is a list of coding-categories ordered by priority."
712 ;; Put coding-categories listed in ARG to L while checking the
713 ;; validity. We assume that `coding-category-list' contains whole
714 ;; coding-categories.
716 (if (null (memq (car arg
) coding-category-list
))
717 (error "Invalid element in argument: %s" (car arg
)))
718 (setq l
(cons (car arg
) l
))
719 (setq arg
(cdr arg
)))
720 ;; Put coding-category not listed in ARG to L.
721 (while coding-category-list
722 (if (null (memq (car coding-category-list
) l
))
723 (setq l
(cons (car coding-category-list
) l
)))
724 (setq coding-category-list
(cdr coding-category-list
)))
725 ;; Update `coding-category-list' and return it.
726 (setq coding-category-list
(nreverse l
))))
730 (defun set-auto-coding (string)
731 "Return coding system for a file which has STRING at the head and tail.
732 STRING is a concatination of the first 1K-byte and
733 the last 3K-byte of the file.
735 It checks for a -*- coding: tag in the first one or two lines of STRING.
736 If there's no coding: tag in the head, it checks local variables spec
737 in the tailing 3K-byte oof STRING.
739 The return value is the specified coding system,
740 or nil if nothing specified.
742 The variable `auto-file-coding-system' (which see) is set to this
743 function by default."
745 (let ((case-fold-search t
)
746 (len (length string
))
747 (limit (string-match "\n" string
))
750 ;; At first check the head.
752 (when (string-match "^#!" string
)
753 ;; If the file begins with "#!" (exec interpreter
754 ;; magic), look for coding frobs in the first two lines.
755 ;; You cannot necessarily put them in the first line of
756 ;; such a file without screwing up the interpreter
758 (setq limit
(string-match "\n" string limit
))
762 (when (and (string-match "-\\*-\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)" string
)
763 (< (match-beginning 2) limit
))
765 (intern (substring string
(match-beginning 2) (match-end 2))))
766 (if (not (coding-system-p coding-system
))
767 (setq coding-system nil
)))
769 ;; If no coding system is specified in the head, check the tail.
770 (when (and (not coding-system
)
771 (let ((idx (if (> len
3000) (- len
3000) 0))
773 (while (setq start
(string-match "\n\^L" string idx
))
774 (setq idx
(+ start
2)))
776 "^\\(.*\\)[ \t]*Local Variables:[ \t]*\\(.*\\)$"
778 ;; The prefix is what comes before "local variables:" in its line.
779 ;; The suffix is what comes after "local variables:" in its line.
780 (let* ((idx (1+ (match-end 0)))
781 (prefix (regexp-quote
783 (match-beginning 1) (match-end 1))))
784 (suffix (regexp-quote
786 (match-beginning 2) (match-end 2))))
787 (re-coding (concat "^" prefix
788 "coding[ \t]*:[ \t]*\\([^ \t]+\\)[ \t]*"
790 (re-end (concat "^" prefix
"end *:[ \t]*" suffix
"$"))
791 (limit (or (string-match re-end string idx
) len
)))
792 (when (and (setq idx
(string-match re-coding string idx
))
795 (intern (substring string
796 (match-beginning 1) (match-end 1))))
797 (or (coding-system-p coding-system
)
798 (setq coding-system nil
)))))
803 (setq set-auto-coding-function
'set-auto-coding
)
805 ;; Set buffer-file-coding-system of the current buffer after some text
807 (defun after-insert-file-set-buffer-file-coding-system (inserted)
808 (if last-coding-system-used
810 (find-new-buffer-file-coding-system last-coding-system-used
))
811 (modified-p (buffer-modified-p)))
813 (set-buffer-file-coding-system coding-system
)
814 (if (or (eq coding-system
'no-conversion
)
815 (eq (coding-system-type coding-system
) 5))
816 ;; It seems that random 8-bit codes are read. We had
817 ;; better edit this buffer without multibyte character
819 (setq enable-multibyte-characters nil
))
820 (set-buffer-modified-p modified-p
))))
823 (setq after-insert-file-functions
824 (cons 'after-insert-file-set-buffer-file-coding-system
825 after-insert-file-functions
))
827 ;; The coding-spec and eol-type of coding-system returned is decided
828 ;; independently in the following order.
829 ;; 1. That of buffer-file-coding-system locally bound.
830 ;; 2. That of CODING.
832 (defun find-new-buffer-file-coding-system (coding)
833 "Return a coding system for a buffer when a file of CODING is inserted.
834 The local variable `buffer-file-coding-system' of the current buffer
835 is set to the returned value.
836 Return nil if there's no need of setting new buffer-file-coding-system."
837 (let (local-coding local-eol
838 found-coding found-eol
841 ;; Nothing found about coding.
844 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
846 (setq local-eol
(coding-system-eol-type buffer-file-coding-system
))
847 (if (null (numberp local-eol
))
848 ;; But eol-type is not yet set.
849 (setq local-eol nil
))
850 (if (and buffer-file-coding-system
851 (not (eq (coding-system-type buffer-file-coding-system
) t
)))
852 ;; This is not `undecided'.
853 (setq local-coding
(coding-system-base buffer-file-coding-system
)))
855 (if (and (local-variable-p 'buffer-file-coding-system
)
856 local-eol local-coding
)
857 ;; The current buffer has already set full coding-system, we
858 ;; had better not change it.
861 (setq found-eol
(coding-system-eol-type coding
))
862 (if (null (numberp found-eol
))
863 ;; But eol-type is not found.
864 (setq found-eol nil
))
865 (if (not (eq (coding-system-type coding
) t
))
866 ;; This is not `undecided'.
867 (setq found-coding
(coding-system-base coding
)))
869 ;; The local setting takes precedence over the found one.
870 (setq new-coding
(or (and (local-variable-p 'buffer-file-coding-system
)
874 (setq new-eol
(or (and (local-variable-p 'buffer-file-coding-system
)
878 (when (numberp new-eol
)
880 (setq new-coding
'undecided
))
881 (if (vectorp (coding-system-eol-type new-coding
))
883 (aref (coding-system-eol-type new-coding
) new-eol
))))
884 ;; Return a new coding system only when it is different from
886 (if (not (eq buffer-file-coding-system new-coding
))
889 (defun modify-coding-system-alist (target-type regexp coding-system
)
890 "Modify one of look up tables for finding a coding system on I/O operation.
891 There are three of such tables, `file-coding-system-alist',
892 `process-coding-system-alist', and `network-coding-system-alist'.
894 TARGET-TYPE specifies which of them to modify.
895 If it is `file', it affects `file-coding-system-alist' (which see).
896 If it is `process', it affects `process-coding-system-alist' (which see).
897 If it is `network', it affects `network-codign-system-alist' (which see).
899 REGEXP is a regular expression matching a target of I/O operation.
900 The target is a file name if TARGET-TYPE is `file', a program name if
901 TARGET-TYPE is `process', or a network service name or a port number
902 to connect to if TARGET-TYPE is `network'.
904 CODING-SYSTEM is a coding system to perform code conversion on the I/O
905 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
906 for decoding and encoding respectively,
907 or a function symbol which, when called, returns such a cons cell."
908 (or (memq target-type
'(file process network
))
909 (error "Invalid target type: %s" target-type
))
911 (and (eq target-type
'network
) (integerp regexp
))
912 (error "Invalid regular expression: %s" regexp
))
913 (if (symbolp coding-system
)
914 (if (not (fboundp coding-system
))
916 (check-coding-system coding-system
)
917 (setq coding-system
(cons coding-system coding-system
))))
918 (check-coding-system (car coding-system
))
919 (check-coding-system (cdr coding-system
)))
920 (cond ((eq target-type
'file
)
921 (let ((slot (assoc regexp file-coding-system-alist
)))
923 (setcdr slot coding-system
)
924 (setq file-coding-system-alist
925 (cons (cons regexp coding-system
)
926 file-coding-system-alist
)))))
927 ((eq target-type
'process
)
928 (let ((slot (assoc regexp process-coding-system-alist
)))
930 (setcdr slot coding-system
)
931 (setq process-coding-system-alist
932 (cons (cons regexp coding-system
)
933 process-coding-system-alist
)))))
935 (let ((slot (assoc regexp network-coding-system-alist
)))
937 (setcdr slot coding-system
)
938 (setq network-coding-system-alist
939 (cons (cons regexp coding-system
)
940 network-coding-system-alist
)))))))
942 (defun make-unification-table (&rest args
)
943 "Make a unification table (char table) from arguments.
944 Each argument is a list of the form (FROM . TO),
945 where FROM is a character to be unified to TO.
947 FROM can be a generic character (see make-char). In this case, TO is
948 a generic character containing the same number of charcters or a
949 oridinal character. If FROM and TO are both generic characters, all
950 characters belonging to FROM are unified to characters belonging to TO
951 without changing their position code(s)."
952 (let ((table (make-char-table 'character-unification-table
))
955 (let ((elts (car args
)))
957 (let* ((from (car (car elts
)))
958 (from-i 0) ; degree of freedom of FROM
959 (from-rev (nreverse (split-char from
)))
960 (to (cdr (car elts
)))
961 (to-i 0) ; degree of freedom of TO
962 (to-rev (nreverse (split-char to
))))
963 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
964 (while (eq (car from-rev
) 0)
965 (setq from-i
(1+ from-i
) from-rev
(cdr from-rev
)))
966 (while (eq (car to-rev
) 0)
967 (setq to-i
(1+ to-i
) to-rev
(cdr to-rev
)))
968 (if (and (/= from-i to-i
) (/= to-i
0))
969 (error "Invalid character pair (%d . %d)" from to
))
970 ;; If we have already unified TO to TO-ALT, FROM should
971 ;; also be unified to TO-ALT. But, this is only if TO is
972 ;; a generic character or TO-ALT is not a generic
974 (let ((to-alt (aref table to
)))
976 (or (> to-i
0) (not (generic-char-p to-alt
))))
979 (set-char-table-default table from to
)
980 (aset table from to
))
981 ;; If we have already unified some chars to FROM, they
982 ;; should also be unified to TO.
983 (let ((l (assq from revlist
)))
990 (setq l
(cdr l
)) ))))
991 ;; Now update REVLIST.
992 (let ((l (assq to revlist
)))
994 (setcdr l
(cons from
(cdr l
)))
995 (setq revlist
(cons (list to from
) revlist
)))))
996 (setq elts
(cdr elts
))))
997 (setq args
(cdr args
)))
998 ;; Return TABLE just created.
1001 ;;; Initialize some variables.
1003 (put 'use-default-ascent
'char-table-extra-slots
0)
1004 (setq use-default-ascent
(make-char-table 'use-default-ascent
))
1005 (put 'ignore-relative-composition
'char-table-extra-slots
0)
1006 (setq ignore-relative-composition
1007 (make-char-table 'ignore-relative-composition
))
1012 ;;; mule.el ends here