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 ;; So that we don't get completely screwed if the
49 ;; file is encoded in some complicated character set,
50 ;; read it with real decoding, as a multibyte buffer,
51 ;; even if this is a --unibyte Emacs session.
52 (let ((default-major-mode 'fundamental-mode
)
53 (default-enable-multibyte-characters t
))
54 ;; We can't use `generate-new-buffer' because files.el
56 (get-buffer-create (generate-new-buffer-name " *load*"))))
58 (source (save-match-data (string-match "\\.el\\'" fullname
))))
61 (message "Loading %s (source)..." file
)
62 (message "Loading %s..." file
)))
64 (setq preloaded-file-list
(cons file preloaded-file-list
)))
66 (let ((load-file-name fullname
)
67 (inhibit-file-name-operation nil
))
70 (insert-file-contents fullname
)
71 ;; Make `kill-buffer' quiet.
72 (set-buffer-modified-p nil
))
73 ;; Have the original buffer current while we eval.
74 (eval-buffer buffer nil file
75 ;; If this Emacs is running with --unibyte,
76 ;; convert multibyte strings to unibyte
77 ;; after reading them.
78 ;; (not default-enable-multibyte-characters)
80 (let (kill-buffer-hook kill-buffer-query-functions
)
81 (kill-buffer buffer
)))
82 (let ((hook (assoc file after-load-alist
)))
84 (mapcar (function eval
) (cdr hook
))))
85 (unless (or nomessage noninteractive
)
87 (message "Loading %s (source)...done" file
)
88 (message "Loading %s...done" file
)))
91 ;; API (Application Program Interface) for charsets.
93 ;; Return t if OBJ is a quoted symbol
94 ;; and the symbol is the name of a standard charset.
95 (defsubst charset-quoted-standard-p
(obj)
96 (and (listp obj
) (eq (car obj
) 'quote
)
97 (symbolp (car-safe (cdr obj
)))
98 (let ((vector (get (car-safe (cdr obj
)) 'charset
)))
100 (< (aref vector
0) 160)))))
102 (defsubst charsetp
(object)
103 "T is OBJECT is a charset."
104 (and (symbolp object
) (vectorp (get object
'charset
))))
106 (defsubst charset-info
(charset)
107 "Return a vector of information of CHARSET.
108 The elements of the vector are:
109 CHARSET-ID, BYTES, DIMENSION, CHARS, WIDTH, DIRECTION,
110 LEADING-CODE-BASE, LEADING-CODE-EXT,
111 ISO-FINAL-CHAR, ISO-GRAPHIC-PLANE,
112 REVERSE-CHARSET, SHORT-NAME, LONG-NAME, DESCRIPTION,
115 CHARSET-ID (integer) is the identification number of the charset.
116 DIMENSION (integer) is the number of bytes to represent a character of
118 CHARS (integer) is the number of characters in a dimension: 94 or 96.
119 BYTE (integer) is the length of multi-byte form of a character in
120 the charset: one of 1, 2, 3, and 4.
121 WIDTH (integer) is the number of columns a character in the charset
122 occupies on the screen: one of 0, 1, and 2.
123 DIRECTION (integer) is the rendering direction of characters in the
124 charset when rendering. If 0, render from right to left, else
125 render from left to right.
126 LEADING-CODE-BASE (integer) is the base leading-code for the
128 LEADING-CODE-EXT (integer) is the extended leading-code for the
129 charset. All charsets of less than 0xA0 has the value 0.
130 ISO-FINAL-CHAR (character) is the final character of the
131 corresponding ISO 2022 charset.
132 ISO-GRAPHIC-PLANE (integer) is the graphic plane to be invoked
133 while encoding to variants of ISO 2022 coding system, one of the
134 following: 0/graphic-plane-left(GL), 1/graphic-plane-right(GR).
135 REVERSE-CHARSET (integer) is the charset which differs only in
136 LEFT-TO-RIGHT value from the charset. If there's no such a
137 charset, the value is -1.
138 SHORT-NAME (string) is the short name to refer to the charset.
139 LONG-NAME (string) is the long name to refer to the charset
140 DESCRIPTION (string) is the description string of the charset.
141 PLIST (property list) may contain any type of information a user
142 want to put and get by functions `put-charset-property' and
143 `get-charset-property' respectively."
144 (get charset
'charset
))
146 (defmacro charset-id
(charset)
147 "Return charset identification number of CHARSET."
148 (if (charset-quoted-standard-p charset
)
149 (aref (charset-info (nth 1 charset
)) 0)
150 `(aref (charset-info ,charset
) 0)))
152 (defmacro charset-bytes
(charset)
153 "Return bytes of CHARSET.
154 See the function `charset-info' for more detail."
155 (if (charset-quoted-standard-p charset
)
156 (aref (charset-info (nth 1 charset
)) 1)
157 `(aref (charset-info ,charset
) 1)))
159 (defmacro charset-dimension
(charset)
160 "Return dimension of CHARSET.
161 See the function `charset-info' for more detail."
162 (if (charset-quoted-standard-p charset
)
163 (aref (charset-info (nth 1 charset
)) 2)
164 `(aref (charset-info ,charset
) 2)))
166 (defmacro charset-chars
(charset)
167 "Return character numbers contained in a dimension of CHARSET.
168 See the function `charset-info' for more detail."
169 (if (charset-quoted-standard-p charset
)
170 (aref (charset-info (nth 1 charset
)) 3)
171 `(aref (charset-info ,charset
) 3)))
173 (defmacro charset-width
(charset)
174 "Return width (how many column occupied on a screen) of CHARSET.
175 See the function `charset-info' for more detail."
176 (if (charset-quoted-standard-p charset
)
177 (aref (charset-info (nth 1 charset
)) 4)
178 `(aref (charset-info ,charset
) 4)))
180 (defmacro charset-direction
(charset)
181 "Return direction of CHARSET.
182 See the function `charset-info' for more detail."
183 (if (charset-quoted-standard-p charset
)
184 (aref (charset-info (nth 1 charset
)) 5)
185 `(aref (charset-info ,charset
) 5)))
187 (defmacro charset-iso-final-char
(charset)
188 "Return final char of CHARSET.
189 See the function `charset-info' for more detail."
190 (if (charset-quoted-standard-p charset
)
191 (aref (charset-info (nth 1 charset
)) 8)
192 `(aref (charset-info ,charset
) 8)))
194 (defmacro charset-iso-graphic-plane
(charset)
195 "Return graphic plane of CHARSET.
196 See the function `charset-info' for more detail."
197 (if (charset-quoted-standard-p charset
)
198 (aref (charset-info (nth 1 charset
)) 9)
199 `(aref (charset-info ,charset
) 9)))
201 (defmacro charset-reverse-charset
(charset)
202 "Return reverse charset of CHARSET.
203 See the function `charset-info' for more detail."
204 (if (charset-quoted-standard-p charset
)
205 (aref (charset-info (nth 1 charset
)) 10)
206 `(aref (charset-info ,charset
) 10)))
208 (defmacro charset-short-name
(charset)
209 "Return short name of CHARSET.
210 See the function `charset-info' for more detail."
211 (if (charset-quoted-standard-p charset
)
212 (aref (charset-info (nth 1 charset
)) 11)
213 `(aref (charset-info ,charset
) 11)))
215 (defmacro charset-long-name
(charset)
216 "Return long name of CHARSET.
217 See the function `charset-info' for more detail."
218 (if (charset-quoted-standard-p charset
)
219 (aref (charset-info (nth 1 charset
)) 12)
220 `(aref (charset-info ,charset
) 12)))
222 (defmacro charset-description
(charset)
223 "Return descriptoin of CHARSET.
224 See the function `charset-info' for more detail."
225 (if (charset-quoted-standard-p charset
)
226 (aref (charset-info (nth 1 charset
)) 13)
227 `(aref (charset-info ,charset
) 13)))
229 (defmacro charset-plist
(charset)
230 "Return list charset property of CHARSET.
231 See the function `charset-info' for more detail."
232 (if (charset-quoted-standard-p charset
)
233 `(aref ,(charset-info (nth 1 charset
)) 14)
234 `(aref (charset-info ,charset
) 14)))
236 (defun set-charset-plist (charset plist
)
237 "Set CHARSET's property list to PLIST, and retrun PLIST."
238 (aset (charset-info charset
) 14 plist
))
240 (defun make-char (charset &optional c1 c2
)
241 "Return a character of CHARSET and position-codes CODE1 and CODE2.
242 CODE1 and CODE2 are optional, but if you don't supply
243 sufficient position-codes, return a generic character which stands for
244 all characters or group of characters in the character sets.
245 A generic character can be used to index a char table (e.g. syntax-table)."
246 (make-char-internal (charset-id charset
) c1 c2
))
248 (put 'make-char
'byte-compile
251 (let ((charset (nth 1 form
)))
252 (if (charset-quoted-standard-p charset
)
253 (byte-compile-normal-call
254 (cons 'make-char-internal
255 (cons (charset-id (nth 1 charset
)) (nthcdr 2 form
))))
256 (byte-compile-normal-call
257 (cons 'make-char-internal
258 (cons (list 'charset-id charset
) (nthcdr 2 form
)))))))))
260 (defun charset-list ()
261 "Return list of charsets ever defined.
263 This function is provided for backward compatibility.
264 Now we have the variable `charset-list'."
267 (make-obsolete 'charset-list
268 "Use the variable charset-list instead.")
270 (defsubst generic-char-p
(char)
271 "Return t if and only if CHAR is a generic character.
272 See also the documentation of make-char."
274 (let ((l (split-char char
)))
275 (and (or (= (nth 1 l
) 0) (eq (nth 2 l
) 0))
276 (not (eq (car l
) 'composition
))))))
279 ;; Coding system staffs
281 ;; Coding system is a symbol that has the property `coding-system'.
283 ;; The value of the property `coding-system' is a vector of the
285 ;; [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
286 ;; We call this vector as coding-spec. See comments in src/coding.c
289 (defconst coding-spec-type-idx
0)
290 (defconst coding-spec-mnemonic-idx
1)
291 (defconst coding-spec-doc-string-idx
2)
292 (defconst coding-spec-plist-idx
3)
293 (defconst coding-spec-flags-idx
4)
295 ;; PLIST is a property list of a coding system. To share PLIST among
296 ;; alias coding systems, a coding system has PLIST in coding-spec
297 ;; instead of having it in normal property list of Lisp symbol.
298 ;; Here's a list of coding system properties currently being used.
302 ;; The value is a coding category the coding system belongs to. The
303 ;; function `make-coding-system' and `define-coding-system-alias' sets
304 ;; this value automatically.
306 ;; o alias-coding-systems
308 ;; The value is a list of coding systems of the same alias group. The
309 ;; first element is the coding system made at first, which we call as
310 ;; `base coding system'. The function `make-coding-system' and
311 ;; `define-coding-system-alias' set this value automatically.
313 ;; o post-read-conversion
315 ;; The value is a function to call after some text is inserted and
316 ;; decoded by the coding system itself and before any functions in
317 ;; `after-insert-functions' are called. The arguments to this
318 ;; function is the same as those of a function in
319 ;; `after-insert-functions', i.e. LENGTH of a text while putting point
320 ;; at the head of the text to be decoded
322 ;; o pre-write-conversion
324 ;; The value is a function to call after all functions in
325 ;; `write-region-annotate-functions' and `buffer-file-format' are
326 ;; called, and before the text is encoded by the coding system itself.
327 ;; The arguments to this function is the same as those of a function
328 ;; in `write-region-annotate-functions', i.e. FROM and TO specifying
331 ;; o character-unification-table-for-decode
333 ;; The value is a unification table to be applied on decoding. See
334 ;; the function `make-unification-table' for the format of unification
337 ;; o character-unification-table-for-encode
339 ;; The value is a unification table to be applied on encoding.
343 ;; The value is a list of charsets safely supported by the coding
344 ;; system. The value t means that all charsets Emacs handles are
345 ;; supported. Even if some charset is not in this list, it doesn't
346 ;; mean that the charset can't be encoded in the coding system,
347 ;; instead, it just means that some other receiver of a text encoded
348 ;; in the coding system won't be able to handle that charset.
351 ;; Return coding-spec of CODING-SYSTEM
352 (defsubst coding-system-spec
(coding-system)
353 (get (check-coding-system coding-system
) 'coding-system
))
355 (defun coding-system-type (coding-system)
356 "Return the coding type of CODING-SYSTEM.
357 A coding type is an integer value indicating the encoding method
358 of CODING-SYSTEM. See the function `make-coding-system' for more detail."
359 (aref (coding-system-spec coding-system
) coding-spec-type-idx
))
361 (defun coding-system-mnemonic (coding-system)
362 "Return the mnemonic character of CODING-SYSTEM.
363 A mnemonic character of a coding system is used in mode line
364 to indicate the coding system."
365 (or (aref (coding-system-spec coding-system
) coding-spec-mnemonic-idx
)
368 (defun coding-system-doc-string (coding-system)
369 "Return the documentation string for CODING-SYSTEM."
370 (aref (coding-system-spec coding-system
) coding-spec-doc-string-idx
))
372 (defun coding-system-plist (coding-system)
373 "Return the property list of CODING-SYSTEM."
374 (aref (coding-system-spec coding-system
) coding-spec-plist-idx
))
376 (defun coding-system-flags (coding-system)
377 "Return `flags' of CODING-SYSTEM.
378 A `flags' of a coding system is a vector of length 32 indicating detailed
379 information of a coding system. See the function `make-coding-system'
381 (aref (coding-system-spec coding-system
) coding-spec-flags-idx
))
383 (defun coding-system-get (coding-system prop
)
384 "Extract a value from CODING-SYSTEM's property list for property PROP."
385 (plist-get (coding-system-plist coding-system
) prop
))
387 (defun coding-system-put (coding-system prop val
)
388 "Change value in CODING-SYSTEM's property list PROP to VAL."
389 (let ((plist (coding-system-plist coding-system
)))
391 (plist-put plist prop val
)
392 (aset (coding-system-spec coding-system
) coding-spec-plist-idx
395 (defun coding-system-category (coding-system)
396 "Return the coding category of CODING-SYSTEM."
397 (coding-system-get coding-system
'coding-category
))
399 (defun coding-system-base (coding-system)
400 "Return the base coding system of CODING-SYSTEM.
401 A base coding system is what made by `make-coding-system'.
402 Any alias nor subsidiary coding systems are not base coding system."
403 (car (coding-system-get coding-system
'alias-coding-systems
)))
405 (defalias 'coding-system-parent
'coding-system-base
)
406 (make-obsolete 'coding-system-parent
'coding-system-base
)
408 ;; Coding system also has a property `eol-type'.
410 ;; This property indicates how the coding system handles end-of-line
411 ;; format. The value is integer 0, 1, 2, or a vector of three coding
412 ;; systems. Each integer value 0, 1, and 2 indicates the format of
413 ;; end-of-line LF, CRLF, and CR respectively. A vector value
414 ;; indicates that the format of end-of-line should be detected
415 ;; automatically. Nth element of the vector is the subsidiary coding
416 ;; system whose `eol-type' property is N.
418 (defun coding-system-eol-type (coding-system)
419 "Return eol-type of CODING-SYSTEM.
420 An eol-type is integer 0, 1, 2, or a vector of coding systems.
422 Integer values 0, 1, and 2 indicate a format of end-of-line; LF,
423 CRLF, and CR respectively.
425 A vector value indicates that a format of end-of-line should be
426 detected automatically. Nth element of the vector is the subsidiary
427 coding system whose eol-type is N."
428 (get coding-system
'eol-type
))
430 ;; Make subsidiary coding systems (eol-type variants) of CODING-SYSTEM.
431 (defun make-subsidiary-coding-system (coding-system)
432 (let ((coding-spec (coding-system-spec coding-system
))
433 (subsidiaries (vector (intern (format "%s-unix" coding-system
))
434 (intern (format "%s-dos" coding-system
))
435 (intern (format "%s-mac" coding-system
))))
439 (put (aref subsidiaries i
) 'coding-system coding-spec
)
440 (put (aref subsidiaries i
) 'eol-type i
)
441 (setq coding-system-list
442 (cons (aref subsidiaries i
) coding-system-list
))
443 (setq coding-system-alist
444 (cons (list (symbol-name (aref subsidiaries i
)))
445 coding-system-alist
))
449 (defun make-coding-system (coding-system type mnemonic doc-string
450 &optional flags properties
)
451 "Define a new coding system CODING-SYSTEM (symbol).
452 Remaining arguments are TYPE, MNEMONIC, DOC-STRING, FLAGS (optional),
453 and PROPERTIES (optional) which construct a coding-spec of CODING-SYSTEM
454 in the following format:
455 [TYPE MNEMONIC DOC-STRING PLIST FLAGS]
457 TYPE is an integer value indicating the type of the coding system as follows:
458 0: Emacs internal format,
459 1: Shift-JIS (or MS-Kanji) used mainly on Japanese PC,
460 2: ISO-2022 including many variants,
461 3: Big5 used mainly on Chinese PC,
462 4: private, CCL programs provide encoding/decoding algorithm,
463 5: Raw-text, which means that text contains random 8-bit codes.
465 MNEMONIC is a character to be displayed on mode line for the coding system.
467 DOC-STRING is a documentation string for the coding system.
469 FLAGS specifies more detailed information of the coding system as follows:
471 If TYPE is 2 (ISO-2022), FLAGS is a list of these elements:
472 CHARSET0, CHARSET1, CHARSET2, CHARSET3, SHORT-FORM,
473 ASCII-EOL, ASCII-CNTL, SEVEN, LOCKING-SHIFT, SINGLE-SHIFT,
474 USE-ROMAN, USE-OLDJIS, NO-ISO6429, INIT-BOL, DESIGNATION-BOL,
475 SAFE, ACCEPT-LATIN-EXTRA-CODE.
476 CHARSETn are character sets initially designated to Gn graphic registers.
477 If CHARSETn is nil, Gn is never used.
478 If CHARSETn is t, Gn can be used but nothing designated initially.
479 If CHARSETn is a list of character sets, those character sets are
480 designated to Gn on output, but nothing designated to Gn initially.
481 SHORT-FORM non-nil means use short designation sequence on output.
482 ASCII-EOL non-nil means designate ASCII to g0 at end of line on output.
483 ASCII-CNTL non-nil means designate ASCII to g0 before control codes and
485 SEVEN non-nil means use 7-bit code only on output.
486 LOCKING-SHIFT non-nil means use locking-shift.
487 SINGLE-SHIFT non-nil means use single-shift.
488 USE-ROMAN non-nil means designate JIS0201-1976-Roman instead of ASCII.
489 USE-OLDJIS non-nil means designate JIS0208-1976 instead of JIS0208-1983.
490 NO-ISO6429 non-nil means not use ISO6429's direction specification.
491 INIT-BOL non-nil means any designation state is assumed to be reset
492 to initial at each beginning of line on output.
493 DESIGNATION-BOL non-nil means designation sequences should be placed
494 at beginning of line on output.
495 SAFE non-nil means convert unsafe characters to `?' on output.
496 Unsafe characters are what not specified in SAFE-CHARSET.
497 ACCEPT-LATIN-EXTRA-CODE non-nil means code-detection routine accepts
498 a code specified in `latin-extra-code-table' (which see) as a valid
499 code of the coding system.
501 If TYPE is 4 (private), FLAGS should be a cons of CCL programs, for
502 decoding and encoding. CCL programs should be specified by their
505 PROPERTIES is an alist of properties vs the corresponding values.
506 These properties are set in PLIST, a property list. This function
507 also sets properties `coding-category' and `alias-coding-systems'
510 Kludgy features for backward compatibility:
512 1. If TYPE is 4 and car or cdr of FLAGS is a vector, the vector is
513 treated as a compiled CCL code.
515 2. If PROPERTIES is just a list of character sets, the list is set as
516 a value of `safe-charsets' in PLIST."
517 (if (memq coding-system coding-system-list
)
518 (error "Coding system %s already exists" coding-system
))
520 ;; Set a value of `coding-system' property.
521 (let ((coding-spec (make-vector 5 nil
))
522 (no-initial-designation t
)
523 (no-alternative-designation t
)
525 (if (or (not (integerp type
)) (< type
0) (> type
5))
526 (error "TYPE argument must be 0..5"))
527 (if (or (not (integerp mnemonic
)) (<= mnemonic ?
) (> mnemonic
127))
528 (error "MNEMONIC arguemnt must be an ASCII printable character."))
529 (aset coding-spec coding-spec-type-idx type
)
530 (aset coding-spec coding-spec-mnemonic-idx mnemonic
)
531 (aset coding-spec coding-spec-doc-string-idx
532 (if (stringp doc-string
) doc-string
""))
534 (setq coding-category
'coding-category-emacs-mule
))
536 (setq coding-category
'coding-category-sjis
))
537 ((= type
2) ; ISO2022
539 (vec (make-vector 32 nil
))
540 (g1-designation nil
))
542 (let ((charset (car flags
)))
543 (if (and no-initial-designation
545 (or (charsetp charset
)
547 (charsetp (car charset
)))))
548 (setq no-initial-designation nil
))
549 (if (charsetp charset
)
550 (if (= i
1) (setq g1-designation charset
))
555 (setq elt
(car tail
))
557 (setq no-alternative-designation nil
)
558 (if (and elt
(not (charsetp elt
)))
559 (error "Invalid charset: %s" elt
)))
560 (setq tail
(cdr tail
)))
561 (setq g1-designation
(car charset
)))
564 (setq no-alternative-designation nil
)
565 (error "Invalid charset: %s" charset
)))))
566 (aset vec i charset
))
567 (setq flags
(cdr flags
) i
(1+ i
)))
568 (while (and (< i
32) flags
)
569 (aset vec i
(car flags
))
570 (setq flags
(cdr flags
) i
(1+ i
)))
571 (aset coding-spec
4 vec
)
572 (setq coding-category
573 (if (aref vec
8) ; Use locking-shift.
574 (or (and (aref vec
7) 'coding-category-iso-7-else
)
575 'coding-category-iso-8-else
)
576 (if (aref vec
7) ; 7-bit only.
577 (if (aref vec
9) ; Use single-shift.
578 'coding-category-iso-7-else
579 (if no-alternative-designation
580 'coding-category-iso-7-tight
581 'coding-category-iso-7
))
582 (if no-initial-designation
583 'coding-category-iso-8-else
584 (if (and (charsetp g1-designation
)
585 (= (charset-dimension g1-designation
) 2))
586 'coding-category-iso-8-2
587 'coding-category-iso-8-1
)))))))
589 (setq coding-category
'coding-category-big5
))
590 ((= type
4) ; private
591 (setq coding-category
'coding-category-binary
)
592 (if (not (consp flags
))
593 (error "Invalid FLAGS argument for TYPE 4 (CCL)")
594 (let ((decoder (check-ccl-program
596 (intern (format "%s-decoder" coding-system
))))
597 (encoder (check-ccl-program
599 (intern (format "%s-encoder" coding-system
)))))
600 (if (and decoder encoder
)
601 (aset coding-spec
4 (cons decoder encoder
))
602 (error "Invalid FLAGS argument for TYPE 4 (CCL)")))))
604 (setq coding-category
'coding-category-raw-text
)))
606 (let ((plist (list 'coding-category coding-category
607 'alias-coding-systems
(list coding-system
))))
608 (if no-initial-designation
609 (plist-put plist
'no-initial-designation t
))
611 (or (eq properties t
)
612 (not (consp (car properties
)))))
613 ;; In the old version, the arg PROPERTIES is a list to be
614 ;; set in PLIST as a value of property `safe-charsets'.
615 (plist-put plist
'safe-charsets properties
)
617 (plist-put plist
(car (car properties
)) (cdr (car properties
)))
618 (setq properties
(cdr properties
))))
619 (aset coding-spec coding-spec-plist-idx plist
))
620 (put coding-system
'coding-system coding-spec
)
621 (put coding-category
'coding-systems
622 (cons coding-system
(get coding-category
'coding-systems
))))
624 ;; Next, set a value of `eol-type' property. The value is a vector
625 ;; of subsidiary coding systems, each corresponds to a coding system
626 ;; for the detected end-of-line format.
627 (put coding-system
'eol-type
628 (if (or (<= type
3) (= type
5))
629 (make-subsidiary-coding-system coding-system
)
632 ;; At last, register CODING-SYSTEM in `coding-system-list' and
633 ;; `coding-system-alist'.
634 (setq coding-system-list
(cons coding-system coding-system-list
))
635 (setq coding-system-alist
(cons (list (symbol-name coding-system
))
636 coding-system-alist
))
639 (defun define-coding-system-alias (alias coding-system
)
640 "Define ALIAS as an alias for coding system CODING-SYSTEM."
641 (put alias
'coding-system
(coding-system-spec coding-system
))
642 (nconc (coding-system-get alias
'alias-coding-systems
) (list alias
))
643 (setq coding-system-list
(cons alias coding-system-list
))
644 (setq coding-system-alist
(cons (list (symbol-name alias
))
645 coding-system-alist
))
646 (let ((eol-type (coding-system-eol-type coding-system
)))
647 (if (vectorp eol-type
)
648 (put alias
'eol-type
(make-subsidiary-coding-system alias
))
649 (put alias
'eol-type eol-type
))))
651 (defun set-buffer-file-coding-system (coding-system &optional force
)
652 "Set the file coding-system of the current buffer to CODING-SYSTEM.
653 This means that when you save the buffer, it will be converted
654 according to CODING-SYSTEM. For a list of possible values of CODING-SYSTEM,
655 use \\[list-coding-systems].
657 If the buffer's previous file coding-system value specifies end-of-line
658 conversion, and CODING-SYSTEM does not specify one, CODING-SYSTEM is
659 merged with the already-specified end-of-line conversion.
660 However, if the optional prefix argument FORCE is non-nil,
661 then CODING-SYSTEM is used exactly as specified."
662 (interactive "zCoding system for visited file (default, nil): \nP")
663 (check-coding-system coding-system
)
665 (let ((x (coding-system-eol-type buffer-file-coding-system
))
666 (y (coding-system-eol-type coding-system
)))
667 (if (and (numberp x
) (>= x
0) (<= x
2) (vectorp y
))
668 (setq coding-system
(aref y x
)))))
669 (setq buffer-file-coding-system coding-system
)
670 (set-buffer-modified-p t
)
671 (force-mode-line-update))
673 (defvar default-terminal-coding-system nil
674 "Default value for the terminal coding system.
675 This is normally set according to the selected language environment.
676 See also the command `set-terminal-coding-system'.")
678 (defun set-terminal-coding-system (coding-system)
679 "Set coding system of your terminal to CODING-SYSTEM.
680 All text output to the terminal will be encoded
681 with the specified coding system.
682 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
683 The default is determined by the selected language environment
684 or by the previous use of this command."
686 (list (let ((default (if (and (not (terminal-coding-system))
687 default-terminal-coding-system
)
688 default-terminal-coding-system
)))
690 (format "Coding system for terminal display (default, %s): "
693 (if (and (not coding-system
)
694 (not (terminal-coding-system)))
695 (setq coding-system default-terminal-coding-system
))
697 (setq default-terminal-coding-system coding-system
))
698 (set-terminal-coding-system-internal coding-system
)
699 (redraw-frame (selected-frame)))
701 (defvar default-keyboard-coding-system nil
702 "Default value of the keyboard coding system.
703 This is normally set according to the selected language environment.
704 See also the command `set-keyboard-coding-system'.")
706 (defun set-keyboard-coding-system (coding-system)
707 "Set coding system for keyboard input to CODING-SYSTEM.
708 In addition, this command enables Encoded-kbd minor mode.
709 \(If CODING-SYSTEM is nil, Encoded-bkd mode is turned off.)
710 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems].
711 The default is determined by the selected language environment
712 or by the previous use of this command."
714 (list (let ((default (if (and (not (keyboard-coding-system))
715 default-keyboard-coding-system
)
716 default-keyboard-coding-system
)))
718 (format "Coding system for keyboard input (default, %s): "
721 (if (and (not coding-system
)
722 (not (keyboard-coding-system)))
723 (setq coding-system default-keyboard-coding-system
))
725 (setq default-keyboard-coding-system coding-system
))
726 (set-keyboard-coding-system-internal coding-system
)
727 (encoded-kbd-mode (if coding-system
1 0)))
729 (defun set-buffer-process-coding-system (decoding encoding
)
730 "Set coding systems for the process associated with the current buffer.
731 DECODING is the coding system to be used to decode input from the process,
732 ENCODING is the coding system to be used to encode output to the process.
734 For a list of possible values of CODING-SYSTEM, use \\[list-coding-systems]."
736 "zCoding-system for process input: \nzCoding-system for process output: ")
737 (let ((proc (get-buffer-process (current-buffer))))
740 (check-coding-system decoding
)
741 (check-coding-system encoding
)
742 (set-process-coding-system proc decoding encoding
)))
743 (force-mode-line-update))
745 (defun set-coding-priority (arg)
746 "Set priority of coding categories according to LIST.
747 LIST is a list of coding categories ordered by priority."
749 (current-list (copy-sequence coding-category-list
)))
750 ;; Check the varidity of ARG while deleting coding categories in
751 ;; ARG from CURRENT-LIST. We assume that CODING-CATEGORY-LIST
752 ;; contains all coding categories.
754 (if (or (null (get (car l
) 'coding-category-index
))
755 (null (memq (car l
) current-list
)))
756 (error "Invalid or duplicated element in argument: %s" arg
))
757 (setq current-list
(delq (car l
) current-list
))
759 ;; Update `coding-category-list' and return it.
760 (setq coding-category-list
(append arg current-list
))))
764 (defun set-auto-coding (string)
765 "Return coding system for a file which has STRING at the head and tail.
766 STRING is a concatination of the first 1K-byte and
767 the last 3K-byte of the file.
769 It checks for a -*- coding: tag in the first one or two lines of STRING.
770 If there's no coding: tag in the head, it checks local variables spec
771 in the tailing 3K-byte oof STRING.
773 The return value is the specified coding system,
774 or nil if nothing specified.
776 The variable `set-auto-coding-function' (which see) is set to this
777 function by default."
779 (let ((case-fold-search t
)
780 (len (length string
))
781 (limit (string-match "\n" string
))
784 ;; At first check the head.
786 (when (string-match "^#!" string
)
787 ;; If the file begins with "#!" (exec interpreter
788 ;; magic), look for coding frobs in the first two lines.
789 ;; You cannot necessarily put them in the first line of
790 ;; such a file without screwing up the interpreter
792 (setq limit
(string-match "\n" string limit
))
796 (when (and (string-match "-\\*-\\(.*;\\)?[ \t]*coding:[ \t]*\\([^ ;]+\\)" string
)
797 (< (match-beginning 2) limit
))
799 (intern (substring string
(match-beginning 2) (match-end 2))))
800 (if (not (coding-system-p coding-system
))
801 (setq coding-system nil
)))
803 ;; If no coding system is specified in the head, check the tail.
804 (when (and (not coding-system
)
805 (let ((idx (if (> len
3000) (- len
3000) 0))
807 (while (setq start
(string-match "\n\^L" string idx
))
808 (setq idx
(+ start
2)))
810 "^\\(.*\\)[ \t]*Local Variables:[ \t]*\\(.*\\)$"
812 ;; The prefix is what comes before "local variables:" in its line.
813 ;; The suffix is what comes after "local variables:" in its line.
814 (let* ((idx (1+ (match-end 0)))
815 (prefix (regexp-quote
817 (match-beginning 1) (match-end 1))))
818 (suffix (regexp-quote
820 (match-beginning 2) (match-end 2))))
821 (re-coding (concat "^" prefix
822 "coding[ \t]*:[ \t]*\\([^ \t]+\\)[ \t]*"
824 (re-end (concat "^" prefix
"end *:[ \t]*" suffix
"$"))
825 (limit (or (string-match re-end string idx
) len
)))
826 (when (and (setq idx
(string-match re-coding string idx
))
829 (intern (substring string
830 (match-beginning 1) (match-end 1))))
831 (or (coding-system-p coding-system
)
832 (setq coding-system nil
)))))
837 (setq set-auto-coding-function
'set-auto-coding
)
839 ;; Set buffer-file-coding-system of the current buffer after some text
841 (defun after-insert-file-set-buffer-file-coding-system (inserted)
842 (if last-coding-system-used
844 (find-new-buffer-file-coding-system last-coding-system-used
))
845 (modified-p (buffer-modified-p)))
847 (set-buffer-file-coding-system coding-system
)
848 (if (or (eq coding-system
'no-conversion
)
849 (eq (coding-system-type coding-system
) 5))
850 ;; It seems that random 8-bit codes are read. We had
851 ;; better edit this buffer without multibyte characters.
852 (set-buffer-multibyte nil
))
853 (set-buffer-modified-p modified-p
))))
856 (add-hook 'after-insert-file-functions
857 'after-insert-file-set-buffer-file-coding-system
)
859 ;; The coding-spec and eol-type of coding-system returned is decided
860 ;; independently in the following order.
861 ;; 1. That of buffer-file-coding-system locally bound.
862 ;; 2. That of CODING.
864 (defun find-new-buffer-file-coding-system (coding)
865 "Return a coding system for a buffer when a file of CODING is inserted.
866 The local variable `buffer-file-coding-system' of the current buffer
867 is set to the returned value.
868 Return nil if there's no need of setting new buffer-file-coding-system."
869 (let (local-coding local-eol
870 found-coding found-eol
873 ;; Nothing found about coding.
876 ;; Get information of `buffer-file-coding-system' in LOCAL-EOL
878 (setq local-eol
(coding-system-eol-type buffer-file-coding-system
))
879 (if (null (numberp local-eol
))
880 ;; But eol-type is not yet set.
881 (setq local-eol nil
))
882 (if (and buffer-file-coding-system
883 (not (eq (coding-system-type buffer-file-coding-system
) t
)))
884 ;; This is not `undecided'.
885 (setq local-coding
(coding-system-base buffer-file-coding-system
)))
887 (if (and (local-variable-p 'buffer-file-coding-system
)
888 local-eol local-coding
)
889 ;; The current buffer has already set full coding-system, we
890 ;; had better not change it.
893 (setq found-eol
(coding-system-eol-type coding
))
894 (if (null (numberp found-eol
))
895 ;; But eol-type is not found.
896 (setq found-eol nil
))
897 (if (not (eq (coding-system-type coding
) t
))
898 ;; This is not `undecided'.
899 (setq found-coding
(coding-system-base coding
)))
901 ;; The local setting takes precedence over the found one.
902 (setq new-coding
(or (and (local-variable-p 'buffer-file-coding-system
)
906 (setq new-eol
(or (and (local-variable-p 'buffer-file-coding-system
)
910 (when (numberp new-eol
)
912 (setq new-coding
'undecided
))
913 (if (vectorp (coding-system-eol-type new-coding
))
915 (aref (coding-system-eol-type new-coding
) new-eol
))))
916 ;; Return a new coding system only when it is different from
918 (if (not (eq buffer-file-coding-system new-coding
))
921 (defun modify-coding-system-alist (target-type regexp coding-system
)
922 "Modify one of look up tables for finding a coding system on I/O operation.
923 There are three of such tables, `file-coding-system-alist',
924 `process-coding-system-alist', and `network-coding-system-alist'.
926 TARGET-TYPE specifies which of them to modify.
927 If it is `file', it affects `file-coding-system-alist' (which see).
928 If it is `process', it affects `process-coding-system-alist' (which see).
929 If it is `network', it affects `network-codign-system-alist' (which see).
931 REGEXP is a regular expression matching a target of I/O operation.
932 The target is a file name if TARGET-TYPE is `file', a program name if
933 TARGET-TYPE is `process', or a network service name or a port number
934 to connect to if TARGET-TYPE is `network'.
936 CODING-SYSTEM is a coding system to perform code conversion on the I/O
937 operation, or a cons cell (DECODING . ENCODING) specifying the coding systems
938 for decoding and encoding respectively,
939 or a function symbol which, when called, returns such a cons cell."
940 (or (memq target-type
'(file process network
))
941 (error "Invalid target type: %s" target-type
))
943 (and (eq target-type
'network
) (integerp regexp
))
944 (error "Invalid regular expression: %s" regexp
))
945 (if (symbolp coding-system
)
946 (if (not (fboundp coding-system
))
948 (check-coding-system coding-system
)
949 (setq coding-system
(cons coding-system coding-system
))))
950 (check-coding-system (car coding-system
))
951 (check-coding-system (cdr coding-system
)))
952 (cond ((eq target-type
'file
)
953 (let ((slot (assoc regexp file-coding-system-alist
)))
955 (setcdr slot coding-system
)
956 (setq file-coding-system-alist
957 (cons (cons regexp coding-system
)
958 file-coding-system-alist
)))))
959 ((eq target-type
'process
)
960 (let ((slot (assoc regexp process-coding-system-alist
)))
962 (setcdr slot coding-system
)
963 (setq process-coding-system-alist
964 (cons (cons regexp coding-system
)
965 process-coding-system-alist
)))))
967 (let ((slot (assoc regexp network-coding-system-alist
)))
969 (setcdr slot coding-system
)
970 (setq network-coding-system-alist
971 (cons (cons regexp coding-system
)
972 network-coding-system-alist
)))))))
974 (defun make-unification-table (&rest args
)
975 "Make a unification table (char table) from arguments.
976 Each argument is a list of the form (FROM . TO),
977 where FROM is a character to be unified to TO.
979 FROM can be a generic character (see make-char). In this case, TO is
980 a generic character containing the same number of charcters or a
981 oridinal character. If FROM and TO are both generic characters, all
982 characters belonging to FROM are unified to characters belonging to TO
983 without changing their position code(s)."
984 (let ((table (make-char-table 'character-unification-table
))
987 (let ((elts (car args
)))
989 (let* ((from (car (car elts
)))
990 (from-i 0) ; degree of freedom of FROM
991 (from-rev (nreverse (split-char from
)))
992 (to (cdr (car elts
)))
993 (to-i 0) ; degree of freedom of TO
994 (to-rev (nreverse (split-char to
))))
995 ;; Check numbers of heading 0s in FROM-REV and TO-REV.
996 (while (eq (car from-rev
) 0)
997 (setq from-i
(1+ from-i
) from-rev
(cdr from-rev
)))
998 (while (eq (car to-rev
) 0)
999 (setq to-i
(1+ to-i
) to-rev
(cdr to-rev
)))
1000 (if (and (/= from-i to-i
) (/= to-i
0))
1001 (error "Invalid character pair (%d . %d)" from to
))
1002 ;; If we have already unified TO to TO-ALT, FROM should
1003 ;; also be unified to TO-ALT. But, this is only if TO is
1004 ;; a generic character or TO-ALT is not a generic
1006 (let ((to-alt (aref table to
)))
1008 (or (> to-i
0) (not (generic-char-p to-alt
))))
1011 (set-char-table-default table from to
)
1012 (aset table from to
))
1013 ;; If we have already unified some chars to FROM, they
1014 ;; should also be unified to TO.
1015 (let ((l (assq from revlist
)))
1022 (setq l
(cdr l
)) ))))
1023 ;; Now update REVLIST.
1024 (let ((l (assq to revlist
)))
1026 (setcdr l
(cons from
(cdr l
)))
1027 (setq revlist
(cons (list to from
) revlist
)))))
1028 (setq elts
(cdr elts
))))
1029 (setq args
(cdr args
)))
1030 ;; Return TABLE just created.
1033 (defun define-character-unification-table (symbol &rest args
)
1034 "define character unification table. This function call make-unification-table,
1035 store a returned table to character-unification-table-vector.
1036 And then set the table as SYMBOL's unification-table property,
1037 the index of the vector as SYMBOL's unification-table-id."
1038 (let ((table (apply 'make-unification-table args
))
1039 (len (length character-unification-table-vector
))
1042 (or (symbolp symbol
)
1043 (signal 'wrong-type-argument symbol
))
1044 (put symbol
'unification-table table
)
1045 (while (and (< id len
)
1046 (if (consp (setq slot
(aref character-unification-table-vector id
)))
1047 (if (eq (car slot
) symbol
) nil t
)
1048 (aset character-unification-table-vector id
(cons symbol table
))
1053 (setq character-unification-table-vector
1054 (vconcat character-unification-table-vector
(make-vector len nil
)))
1055 (aset character-unification-table-vector id
(cons symbol table
))))
1056 (put symbol
'unification-table-id id
)
1060 ;;; Initialize some variables.
1062 (put 'use-default-ascent
'char-table-extra-slots
0)
1063 (setq use-default-ascent
(make-char-table 'use-default-ascent
))
1064 (put 'ignore-relative-composition
'char-table-extra-slots
0)
1065 (setq ignore-relative-composition
1066 (make-char-table 'ignore-relative-composition
))
1071 ;;; mule.el ends here