1 ;;; mm-util.el --- Utility functions for Mule and low level things
3 ;; Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004,
4 ;; 2005, 2006 Free Software Foundation, Inc.
6 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
7 ;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
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., 51 Franklin Street, Fifth Floor,
23 ;; Boston, MA 02110-1301, USA.
29 (eval-when-compile (require 'cl))
33 (if (featurep 'xemacs)
34 (unless (ignore-errors
35 (require 'timer-funcs))
42 (let ((nfunc (intern (format "mm-%s" (car elem)))))
43 (if (fboundp (car elem))
44 (defalias nfunc (car elem))
45 (defalias nfunc (cdr elem)))))
46 '((decode-coding-string . (lambda (s a) s))
47 (encode-coding-string . (lambda (s a) s))
48 (encode-coding-region . ignore)
49 (coding-system-list . ignore)
50 (decode-coding-region . ignore)
52 (coding-system-equal . equal)
53 (annotationp . ignore)
54 (set-buffer-file-coding-system . ignore)
61 (mapcar (lambda (e) (list (symbol-name (car e))))
62 mm-mime-mule-charset-alist)
65 . (lambda (from to string &optional inplace)
66 ;; stolen (and renamed) from nnheader.el
67 "Replace characters in STRING from FROM to TO.
68 Unless optional argument INPLACE is non-nil, return a new string."
69 (let ((string (if inplace string (copy-sequence string)))
72 ;; Replace all occurrences of FROM with TO.
74 (when (= (aref string idx) from)
79 . (lambda (string regexp rep &optional literal)
80 "See `replace-regexp-in-string', only the order of args differs."
81 (replace-regexp-in-string regexp rep string nil literal)))
82 (string-as-unibyte . identity)
83 (string-make-unibyte . identity)
84 ;; string-as-multibyte often doesn't really do what you think it does.
86 ;; (aref (string-as-multibyte "\201") 0) -> 129 (aka ?\201)
87 ;; (aref (string-as-multibyte "\300") 0) -> 192 (aka ?\300)
88 ;; (aref (string-as-multibyte "\300\201") 0) -> 192 (aka ?\300)
89 ;; (aref (string-as-multibyte "\300\201") 1) -> 129 (aka ?\201)
91 ;; (aref (string-as-multibyte "\201\300") 0) -> 2240
92 ;; (aref (string-as-multibyte "\201\300") 1) -> <error>
93 ;; Better use string-to-multibyte or encode-coding-string.
94 ;; If you really need string-as-multibyte somewhere it's usually
95 ;; because you're using the internal emacs-mule representation (maybe
96 ;; because you're using string-as-unibyte somewhere), which is
97 ;; generally a problem in itself.
98 ;; Here is an approximate equivalence table to help think about it:
99 ;; (string-as-multibyte s) ~= (decode-coding-string s 'emacs-mule)
100 ;; (string-to-multibyte s) ~= (decode-coding-string s 'binary)
101 ;; (string-make-multibyte s) ~= (decode-coding-string s locale-coding-system)
102 (string-as-multibyte . identity)
105 "Return a multibyte string with the same individual chars as string."
107 (lambda (ch) (mm-string-as-multibyte (char-to-string ch)))
109 (multibyte-string-p . ignore)
110 (insert-byte . insert-char)
111 (multibyte-char-to-unibyte . identity)
113 . (lambda (buffer-name)
114 "Returns non-nil if a buffer named BUFFER-NAME gets a special frame."
115 (and special-display-function
116 (or (and (member buffer-name special-display-buffer-names) t)
117 (cdr (assoc buffer-name special-display-buffer-names))
119 (dolist (elem special-display-regexps)
121 (string-match elem buffer-name)
125 (string-match (car elem) buffer-name)
126 (throw 'return (cdr elem))))))))))))
129 (defalias 'mm-char-or-char-int-p
131 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
132 ((fboundp 'char-valid-p) 'char-valid-p)
135 ;; Fixme: This seems always to be used to read a MIME charset, so it
136 ;; should be re-named and fixed (in Emacs) to offer completion only on
137 ;; proper charset names (base coding systems which have a
138 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
140 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
141 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
142 ;; Actually, there should be an `mm-coding-system-mime-charset'.
144 (defalias 'mm-read-coding-system
146 ((fboundp 'read-coding-system)
147 (if (and (featurep 'xemacs)
148 (<= (string-to-number emacs-version) 21.1))
149 (lambda (prompt &optional default-coding-system)
150 (read-coding-system prompt))
151 'read-coding-system))
152 (t (lambda (prompt &optional default-coding-system)
153 "Prompt the user for a coding system."
155 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
156 mm-mime-mule-charset-alist)))))))
158 (defvar mm-coding-system-list nil)
159 (defun mm-get-coding-system-list ()
160 "Get the coding system list."
161 (or mm-coding-system-list
162 (setq mm-coding-system-list (mm-coding-system-list))))
164 (defun mm-coding-system-p (cs)
165 "Return non-nil if CS is a symbol naming a coding system.
166 In XEmacs, also return non-nil if CS is a coding system object.
167 If CS is available, return CS itself in Emacs, and return a coding
168 system object in XEmacs."
169 (if (fboundp 'find-coding-system)
170 (and cs (find-coding-system cs))
171 (if (fboundp 'coding-system-p)
172 (when (coding-system-p cs)
175 (car (memq cs (mm-get-coding-system-list))))))
177 (defun mm-codepage-setup (number &optional alias)
178 "Create a coding system cpNUMBER.
179 The coding system is created using `codepage-setup'. If ALIAS is
180 non-nil, an alias is created and added to
181 `mm-charset-synonym-alist'. If ALIAS is a string, it's used as
182 the alias. Else windows-NUMBER is used."
184 (let ((completion-ignore-case t)
185 (candidates (cp-supported-codepages)))
186 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
187 nil t nil nil "437"))))
189 (setq alias (if (stringp alias)
191 (intern (format "windows-%s" number)))))
192 (let* ((cp (intern (format "cp%s" number))))
193 (unless (mm-coding-system-p cp)
194 (codepage-setup number))
196 ;; Don't add alias if setup of cp failed.
197 (mm-coding-system-p cp))
198 (add-to-list 'mm-charset-synonym-alist (cons alias cp)))))
200 (defvar mm-charset-synonym-alist
202 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
203 ,@(unless (mm-coding-system-p 'x-ctext)
204 '((x-ctext . ctext)))
205 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_!
206 ,@(unless (mm-coding-system-p 'iso-8859-15)
207 '((iso-8859-15 . iso-8859-1)))
208 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
209 ,@(unless (mm-coding-system-p 'big5-hkscs)
210 '((big5-hkscs . big5)))
211 ;; Windows-1252 is actually a superset of Latin-1. See also
212 ;; `gnus-article-dumbquotes-map'.
213 ,@(unless (mm-coding-system-p 'windows-1252)
214 (if (mm-coding-system-p 'cp1252)
215 '((windows-1252 . cp1252))
216 '((windows-1252 . iso-8859-1))))
217 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
218 ;; Outlook users in Czech republic. Use this to allow reading of their
219 ;; e-mails. cp1250 should be defined by M-x codepage-setup.
220 ,@(if (and (not (mm-coding-system-p 'windows-1250))
221 (mm-coding-system-p 'cp1250))
222 '((windows-1250 . cp1250)))
223 ;; A Microsoft misunderstanding.
224 ,@(if (and (not (mm-coding-system-p 'unicode))
225 (mm-coding-system-p 'utf-16-le))
226 '((unicode . utf-16-le)))
227 ;; A Microsoft misunderstanding.
228 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
229 (if (mm-coding-system-p 'cp949)
230 '((ks_c_5601-1987 . cp949))
231 '((ks_c_5601-1987 . euc-kr))))
233 "A mapping from unknown or invalid charset names to the real charset names.")
235 (defcustom mm-charset-override-alist
236 `((iso-8859-1 . windows-1252))
237 "A mapping from undesired charset names to their replacement.
239 You may add pairs like (iso-8859-1 . windows-1252) here,
240 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
241 superset of iso-8859-1."
242 :type '(list (set :inline t
243 (const (iso-8859-1 . windows-1252))
244 (const (undecided . windows-1252)))
247 (cons (symbol :tag "From charset")
248 (symbol :tag "To charset"))))
249 :version "23.0" ;; No Gnus
252 (defcustom mm-charset-eval-alist
253 (if (featurep 'xemacs)
254 nil ;; I don't know what would be useful for XEmacs.
255 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
256 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
257 (windows-1250 . (mm-codepage-setup 1250 t))
258 (windows-1251 . (mm-codepage-setup 1251 t))
259 (windows-1253 . (mm-codepage-setup 1253 t))
260 (windows-1257 . (mm-codepage-setup 1257 t))))
261 "An alist of (CHARSET . FORM) pairs.
262 If an article is encoded in an unknown CHARSET, FORM is
263 evaluated. This allows to load additional libraries providing
264 charsets on demand. If supported by your Emacs version, you
265 could use `autoload-coding-system' here."
266 :version "23.0" ;; No Gnus
267 :type '(list (set :inline t
268 (const (windows-1250 . (mm-codepage-setup 1250 t)))
269 (const (windows-1251 . (mm-codepage-setup 1251 t)))
270 (const (windows-1253 . (mm-codepage-setup 1253 t)))
271 (const (windows-1257 . (mm-codepage-setup 1257 t)))
272 (const (cp850 . (mm-codepage-setup 850 nil))))
275 (cons (symbol :tag "charset")
276 (symbol :tag "form"))))
279 (defvar mm-binary-coding-system
281 ((mm-coding-system-p 'binary) 'binary)
282 ((mm-coding-system-p 'no-conversion) 'no-conversion)
284 "100% binary coding system.")
286 (defvar mm-text-coding-system
287 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
288 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
289 (and (mm-coding-system-p 'raw-text) 'raw-text))
290 mm-binary-coding-system)
291 "Text-safe coding system (For removing ^M).")
293 (defvar mm-text-coding-system-for-write nil
294 "Text coding system for write.")
296 (defvar mm-auto-save-coding-system
298 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
299 (if (memq system-type '(windows-nt ms-dos ms-windows))
300 (if (mm-coding-system-p 'utf-8-emacs-dos)
301 'utf-8-emacs-dos mm-binary-coding-system)
303 ((mm-coding-system-p 'emacs-mule)
304 (if (memq system-type '(windows-nt ms-dos ms-windows))
305 (if (mm-coding-system-p 'emacs-mule-dos)
306 'emacs-mule-dos mm-binary-coding-system)
308 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
309 (t mm-binary-coding-system))
310 "Coding system of auto save file.")
312 (defvar mm-universal-coding-system mm-auto-save-coding-system
313 "The universal coding system.")
315 ;; Fixme: some of the cars here aren't valid MIME charsets. That
316 ;; should only matter with XEmacs, though.
317 (defvar mm-mime-mule-charset-alist
319 (iso-8859-1 latin-iso8859-1)
320 (iso-8859-2 latin-iso8859-2)
321 (iso-8859-3 latin-iso8859-3)
322 (iso-8859-4 latin-iso8859-4)
323 (iso-8859-5 cyrillic-iso8859-5)
324 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
325 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
326 ;; charset is koi8-r, not iso-8859-5.
327 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
328 (iso-8859-6 arabic-iso8859-6)
329 (iso-8859-7 greek-iso8859-7)
330 (iso-8859-8 hebrew-iso8859-8)
331 (iso-8859-9 latin-iso8859-9)
332 (iso-8859-14 latin-iso8859-14)
333 (iso-8859-15 latin-iso8859-15)
334 (viscii vietnamese-viscii-lower)
335 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
336 (euc-kr korean-ksc5601)
337 (gb2312 chinese-gb2312)
338 (big5 chinese-big5-1 chinese-big5-2)
340 (thai-tis620 thai-tis620)
341 (windows-1251 cyrillic-iso8859-5)
342 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
343 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
344 latin-jisx0201 japanese-jisx0208-1978
345 chinese-gb2312 japanese-jisx0208
346 korean-ksc5601 japanese-jisx0212)
347 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
348 latin-jisx0201 japanese-jisx0208-1978
349 chinese-gb2312 japanese-jisx0208
350 korean-ksc5601 japanese-jisx0212
351 chinese-cns11643-1 chinese-cns11643-2)
352 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
353 cyrillic-iso8859-5 greek-iso8859-7
354 latin-jisx0201 japanese-jisx0208-1978
355 chinese-gb2312 japanese-jisx0208
356 korean-ksc5601 japanese-jisx0212
357 chinese-cns11643-1 chinese-cns11643-2
358 chinese-cns11643-3 chinese-cns11643-4
359 chinese-cns11643-5 chinese-cns11643-6
361 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
362 japanese-jisx0213-1 japanese-jisx0213-2)
363 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
364 ,(cond ((fboundp 'unicode-precedence-list)
365 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
366 (unicode-precedence-list)))))
367 ((or (not (fboundp 'charsetp)) ;; non-Mule case
368 (charsetp 'unicode-a)
369 (not (mm-coding-system-p 'mule-utf-8)))
370 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
371 (t ;; If we have utf-8 we're in Mule 5+.
374 (coding-system-get 'mule-utf-8 'safe-charsets))))))
375 "Alist of MIME-charset/MULE-charsets.")
377 (defun mm-enrich-utf-8-by-mule-ucs ()
378 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
379 This function will run when the `un-define' module is loaded under
380 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
381 with Mule charsets. It is completely useless for Emacs."
382 (when (boundp 'unicode-basic-translation-charset-order-list)
388 'unicode-basic-translation-charset-order-list))))
389 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
392 (setq mm-mime-mule-charset-alist
393 (nconc mm-mime-mule-charset-alist
394 (list (cons 'utf-8 val))))))
397 ;; Correct by construction, but should be unnecessary for Emacs:
398 (if (featurep 'xemacs)
399 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
400 (when (and (fboundp 'coding-system-list)
401 (fboundp 'sort-coding-systems))
402 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
406 mime (or (coding-system-get cs :mime-charset) ; Emacs 23 (unicode)
407 (coding-system-get cs 'mime-charset)))
409 (not (eq t (setq mule
410 (coding-system-get cs 'safe-charsets))))
411 (not (assq mime alist)))
412 (push (cons mime (delq 'ascii mule)) alist)))
413 (setq mm-mime-mule-charset-alist (nreverse alist)))))
415 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
416 "A list of special charsets.
417 Valid elements include:
418 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
419 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
422 (defvar mm-iso-8859-15-compatible
423 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
424 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
425 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
427 (defvar mm-iso-8859-x-to-15-table
428 (and (fboundp 'coding-system-p)
429 (mm-coding-system-p 'iso-8859-15)
432 (if (mm-coding-system-p (car cs))
433 (let ((c (string-to-char
434 (decode-coding-string "\341" (car cs)))))
435 (cons (char-charset c)
438 (decode-coding-string "\341" 'iso-8859-15)) c)
439 (string-to-list (decode-coding-string (car (cdr cs))
442 mm-iso-8859-15-compatible))
443 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
445 (defcustom mm-coding-system-priorities
446 (if (boundp 'current-language-environment)
447 (let ((lang (symbol-value 'current-language-environment)))
448 (cond ((string= lang "Japanese")
449 ;; Japanese users prefer iso-2022-jp to euc-japan or
450 ;; shift_jis, however iso-8859-1 should be used when
451 ;; there are only ASCII text and Latin-1 characters.
452 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
453 "Preferred coding systems for encoding outgoing messages.
455 More than one suitable coding system may be found for some text.
456 By default, the coding system with the highest priority is used
457 to encode outgoing messages (see `sort-coding-systems'). If this
458 variable is set, it overrides the default priority."
460 :type '(repeat (symbol :tag "Coding system"))
464 (defvar mm-use-find-coding-systems-region
465 (fboundp 'find-coding-systems-region)
466 "Use `find-coding-systems-region' to find proper coding systems.
468 Setting it to nil is useful on Emacsen supporting Unicode if sending
469 mail with multiple parts is preferred to sending a Unicode one.")
471 ;;; Internal variables:
475 (defun mm-mule-charset-to-mime-charset (charset)
476 "Return the MIME charset corresponding to the given Mule CHARSET."
477 (if (and (fboundp 'find-coding-systems-for-charsets)
478 (fboundp 'sort-coding-systems))
479 (let ((css (sort (sort-coding-systems
480 (find-coding-systems-for-charsets (list charset)))
481 'mm-sort-coding-systems-predicate))
483 (while (and (not mime)
485 (when (setq cs (pop css))
486 (setq mime (or (coding-system-get cs :mime-charset)
487 (coding-system-get cs 'mime-charset)))))
489 (let ((alist (mapcar (lambda (cs)
490 (assq cs mm-mime-mule-charset-alist))
491 (sort (mapcar 'car mm-mime-mule-charset-alist)
492 'mm-sort-coding-systems-predicate)))
495 (when (memq charset (cdar alist))
496 (setq out (caar alist)
501 (defun mm-charset-to-coding-system (charset &optional lbt
503 "Return coding-system corresponding to CHARSET.
504 CHARSET is a symbol naming a MIME charset.
505 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
506 used as the line break code type of the coding system.
508 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
509 map undesired charset names to their replacement. This should
510 only be used for decoding, not for encoding."
511 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
512 (when (stringp charset)
513 (setq charset (intern (downcase charset))))
515 (setq charset (intern (format "%s-%s" charset lbt))))
519 ;; Running in a non-MULE environment.
520 ((or (null (mm-get-coding-system-list))
521 (not (fboundp 'coding-system-get)))
523 ;; Check override list quite early. Should only used for decoding, not for
526 (let ((cs (cdr (assq charset mm-charset-override-alist))))
527 (and cs (mm-coding-system-p cs) cs))))
529 ((eq charset 'us-ascii)
531 ;; Check to see whether we can handle this charset. (This depends
532 ;; on there being some coding system matching each `mime-charset'
533 ;; property defined, as there should be.)
534 ((and (mm-coding-system-p charset)
535 ;;; Doing this would potentially weed out incorrect charsets.
537 ;;; (eq charset (coding-system-get charset 'mime-charset))
540 ;; Eval expressions from `mm-charset-eval-alist'
541 ((let* ((el (assq charset mm-charset-eval-alist))
548 (condition-case nil (eval form) (error nil))
549 ;; (message "Failed to eval `%s'" form))
550 (mm-coding-system-p cs)
551 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
553 ;; Translate invalid charsets.
554 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
556 (mm-coding-system-p cs)
558 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
561 ;; Last resort: search the coding system list for entries which
562 ;; have the right mime-charset in case the canonical name isn't
563 ;; defined (though it should be).
565 ;; mm-get-coding-system-list returns a list of cs without lbt.
567 (dolist (c (mm-get-coding-system-list))
569 (eq charset (or (coding-system-get c :mime-charset)
570 (coding-system-get c 'mime-charset))))
573 ;; Warn the user about unknown charset:
574 (if (fboundp 'gnus-message)
575 (gnus-message 7 "Unknown charset: %s" charset)
576 (message "Unknown charset: %s" charset)))
580 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
581 (boundp 'default-enable-multibyte-characters)
582 default-enable-multibyte-characters
583 (fboundp 'set-buffer-multibyte))
584 "True in Emacs with Mule.")
587 (defun mm-enable-multibyte ()
588 "Set the multibyte flag of the current buffer.
589 Only do this if the default value of `enable-multibyte-characters' is
590 non-nil. This is a no-op in XEmacs."
591 (set-buffer-multibyte 'to))
592 (defalias 'mm-enable-multibyte 'ignore))
595 (defun mm-disable-multibyte ()
596 "Unset the multibyte flag of in the current buffer.
597 This is a no-op in XEmacs."
598 (set-buffer-multibyte nil))
599 (defalias 'mm-disable-multibyte 'ignore)))
601 (defun mm-preferred-coding-system (charset)
602 ;; A typo in some Emacs versions.
603 (or (get-charset-property charset 'preferred-coding-system)
604 (get-charset-property charset 'prefered-coding-system)))
606 ;; Mule charsets shouldn't be used.
607 (defsubst mm-guess-charset ()
608 "Guess Mule charset from the language environment."
610 mail-parse-mule-charset ;; cached mule-charset
612 (setq mail-parse-mule-charset
613 (and (boundp 'current-language-environment)
616 (assoc current-language-environment
617 language-info-alist))))))
618 (if (or (not mail-parse-mule-charset)
619 (eq mail-parse-mule-charset 'ascii))
620 (setq mail-parse-mule-charset
621 (or (car (last (assq mail-parse-charset
622 mm-mime-mule-charset-alist)))
625 mail-parse-mule-charset)))
627 (defun mm-charset-after (&optional pos)
628 "Return charset of a character in current buffer at position POS.
629 If POS is nil, it defauls to the current point.
630 If POS is out of range, the value is nil.
631 If the charset is `composition', return the actual one."
632 (let ((char (char-after pos)) charset)
633 (if (< (mm-char-int char) 128)
634 (setq charset 'ascii)
635 ;; charset-after is fake in some Emacsen.
636 (setq charset (and (fboundp 'char-charset) (char-charset char)))
637 (if (eq charset 'composition) ; Mule 4
638 (let ((p (or pos (point))))
639 (cadr (find-charset-region p (1+ p))))
640 (if (and charset (not (memq charset '(ascii eight-bit-control
641 eight-bit-graphic))))
643 (mm-guess-charset))))))
645 (defun mm-mime-charset (charset)
646 "Return the MIME charset corresponding to the given Mule CHARSET."
647 (if (eq charset 'unknown)
648 (error "The message contains non-printable characters, please use attachment"))
649 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
650 ;; This exists in Emacs 20.
652 (and (mm-preferred-coding-system charset)
653 (or (coding-system-get
654 (mm-preferred-coding-system charset) :mime-charset)
656 (mm-preferred-coding-system charset) 'mime-charset)))
657 (and (eq charset 'ascii)
659 (mm-preferred-coding-system charset)
660 (mm-mule-charset-to-mime-charset charset))
661 ;; This is for XEmacs.
662 (mm-mule-charset-to-mime-charset charset)))
664 (if (fboundp 'delete-dups)
665 (defalias 'mm-delete-duplicates 'delete-dups)
666 (defun mm-delete-duplicates (list)
667 "Destructively remove `equal' duplicates from LIST.
668 Store the result in LIST and return it. LIST must be a proper list.
669 Of several `equal' occurrences of an element in LIST, the first
672 This is a compatibility function for Emacsen without `delete-dups'."
673 ;; Code from `subr.el' in Emacs 22:
676 (setcdr tail (delete (car tail) (cdr tail)))
677 (setq tail (cdr tail))))
680 ;; Fixme: This is used in places when it should be testing the
681 ;; default multibyteness. See mm-default-multibyte-p.
683 (if (and (not (featurep 'xemacs))
684 (boundp 'enable-multibyte-characters))
685 (defun mm-multibyte-p ()
686 "Non-nil if multibyte is enabled in the current buffer."
687 enable-multibyte-characters)
688 (defun mm-multibyte-p () (featurep 'mule))))
690 (defun mm-default-multibyte-p ()
691 "Return non-nil if the session is multibyte.
692 This affects whether coding conversion should be attempted generally."
694 (if (boundp 'default-enable-multibyte-characters)
695 default-enable-multibyte-characters
698 (defun mm-iso-8859-x-to-15-region (&optional b e)
699 (if (fboundp 'char-charset)
700 (let (charset item c inconvertible)
702 (if e (narrow-to-region b e))
703 (goto-char (point-min))
704 (skip-chars-forward "\0-\177")
707 ((not (setq item (assq (char-charset (setq c (char-after)))
708 mm-iso-8859-x-to-15-table)))
710 ((memq c (cdr (cdr item)))
711 (setq inconvertible t)
714 (insert-before-markers (prog1 (+ c (car (cdr item)))
716 (skip-chars-forward "\0-\177")))
717 (not inconvertible))))
719 (defun mm-sort-coding-systems-predicate (a b)
722 ;; Note: invalid entries are dropped silently
723 (and (setq cs (mm-coding-system-p cs))
724 (coding-system-base cs)))
725 mm-coding-system-priorities)))
726 (and (setq a (mm-coding-system-p a))
727 (if (setq b (mm-coding-system-p b))
728 (> (length (memq (coding-system-base a) priorities))
729 (length (memq (coding-system-base b) priorities)))
733 (autoload 'latin-unity-massage-name "latin-unity")
734 (autoload 'latin-unity-maybe-remap "latin-unity")
735 (autoload 'latin-unity-representations-feasible-region "latin-unity")
736 (autoload 'latin-unity-representations-present-region "latin-unity")
737 (defvar latin-unity-coding-systems)
738 (defvar latin-unity-ucs-list))
740 (defun mm-xemacs-find-mime-charset-1 (begin end)
741 "Determine which MIME charset to use to send region as message.
742 This uses the XEmacs-specific latin-unity package to better handle the
743 case where identical characters from diverse ISO-8859-? character sets
744 can be encoded using a single one of the corresponding coding systems.
746 It treats `mm-coding-system-priorities' as the list of preferred
747 coding systems; a useful example setting for this list in Western
748 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
749 to the very standard Latin 1 coding system, and only move to coding
750 systems that are less supported as is necessary to encode the
751 characters that exist in the buffer.
753 Latin Unity doesn't know about those non-ASCII Roman characters that
754 are available in various East Asian character sets. As such, its
755 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
756 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
757 But this is very much a corner case, so don't worry about it."
758 (let ((systems mm-coding-system-priorities) csets psets curset)
760 ;; Load the Latin Unity library, if available.
761 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
762 (require 'latin-unity))
764 ;; Now, can we use it?
765 (if (featurep 'latin-unity)
767 (setq csets (latin-unity-representations-feasible-region begin end)
768 psets (latin-unity-representations-present-region begin end))
772 ;; Pass back the first coding system in the preferred list
773 ;; that can encode the whole region.
774 (dolist (curset systems)
775 (setq curset (latin-unity-massage-name 'buffer-default curset))
777 ;; If the coding system is a universal coding system, then
778 ;; it can certainly encode all the characters in the region.
779 (if (memq curset latin-unity-ucs-list)
780 (throw 'done (list curset)))
782 ;; If a coding system isn't universal, and isn't in
783 ;; the list that latin unity knows about, we can't
784 ;; decide whether to use it here. Leave that until later
785 ;; in `mm-find-mime-charset-region' function, whence we
787 (unless (memq curset latin-unity-coding-systems)
790 ;; Right, we know about this coding system, and it may
791 ;; conceivably be able to encode all the characters in
793 (if (latin-unity-maybe-remap begin end curset csets psets t)
794 (throw 'done (list curset))))
796 ;; Can't encode using anything from the
797 ;; `mm-coding-system-priorities' list.
798 ;; Leave `mm-find-mime-charset' to do most of the work.
801 ;; Right, latin unity isn't available; let `mm-find-charset-region'
802 ;; take its default action, which equally applies to GNU Emacs.
805 (defmacro mm-xemacs-find-mime-charset (begin end)
806 (when (featurep 'xemacs)
807 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
809 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
810 "Return the MIME charsets needed to encode the region between B and E.
811 nil means ASCII, a single-element list represents an appropriate MIME
812 charset, and a longer list means no appropriate charset."
814 ;; The return possibilities of this function are a mess...
815 (or (and (mm-multibyte-p)
816 mm-use-find-coding-systems-region
817 ;; Find the mime-charset of the most preferred coding
818 ;; system that has one.
819 (let ((systems (find-coding-systems-region b e)))
820 (when mm-coding-system-priorities
822 (sort systems 'mm-sort-coding-systems-predicate)))
823 (setq systems (delq 'compound-text systems))
824 (unless (equal systems '(undecided))
826 (let* ((head (pop systems))
827 (cs (or (coding-system-get head :mime-charset)
828 (coding-system-get head 'mime-charset))))
829 ;; The mime-charset (`x-ctext') of
830 ;; `compound-text' is not in the IANA list. We
831 ;; shouldn't normally use anything here with a
832 ;; mime-charset having an `x-' prefix.
833 ;; Fixme: Allow this to be overridden, since
834 ;; there is existing use of x-ctext.
835 ;; Also people apparently need the coding system
836 ;; `iso-2022-jp-3' (which Mule-UCS defines with
837 ;; mime-charset, though it's not valid).
839 (not (string-match "^[Xx]-" (symbol-name cs)))
840 ;; UTF-16 of any variety is invalid for
841 ;; text parts and, unfortunately, has
842 ;; mime-charset defined both in Mule-UCS
843 ;; and versions of Emacs. (The name
844 ;; might be `mule-utf-16...' or
846 (not (string-match "utf-16" (symbol-name cs))))
848 charsets (list cs))))))
850 ;; If we're XEmacs, and some coding system is appropriate,
851 ;; mm-xemacs-find-mime-charset will return an appropriate list.
852 ;; Otherwise, we'll get nil, and the next setq will get invoked.
853 (setq charsets (mm-xemacs-find-mime-charset b e))
855 ;; We're not multibyte, or a single coding system won't cover it.
857 (mm-delete-duplicates
858 (mapcar 'mm-mime-charset
860 (mm-find-charset-region b e))))))
861 (if (and (> (length charsets) 1)
862 (memq 'iso-8859-15 charsets)
863 (memq 'iso-8859-15 hack-charsets)
864 (save-excursion (mm-iso-8859-x-to-15-region b e)))
865 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
866 mm-iso-8859-15-compatible))
867 (if (and (memq 'iso-2022-jp-2 charsets)
868 (memq 'iso-2022-jp-2 hack-charsets))
869 (setq charsets (delq 'iso-2022-jp charsets)))
870 ;; Attempt to reduce the number of charsets if utf-8 is available.
871 (if (and (featurep 'xemacs)
872 (> (length charsets) 1)
873 (mm-coding-system-p 'utf-8))
874 (let ((mm-coding-system-priorities
875 (cons 'utf-8 mm-coding-system-priorities)))
877 (mm-delete-duplicates
878 (mapcar 'mm-mime-charset
880 (mm-find-charset-region b e)))))))
883 (defmacro mm-with-unibyte-buffer (&rest forms)
884 "Create a temporary buffer, and evaluate FORMS there like `progn'.
885 Use unibyte mode for this."
886 `(let (default-enable-multibyte-characters)
887 (with-temp-buffer ,@forms)))
888 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
889 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
891 (defmacro mm-with-multibyte-buffer (&rest forms)
892 "Create a temporary buffer, and evaluate FORMS there like `progn'.
893 Use multibyte mode for this."
894 `(let ((default-enable-multibyte-characters t))
895 (with-temp-buffer ,@forms)))
896 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
897 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
899 (defmacro mm-with-unibyte-current-buffer (&rest forms)
900 "Evaluate FORMS with current buffer temporarily made unibyte.
901 Also bind `default-enable-multibyte-characters' to nil.
902 Equivalent to `progn' in XEmacs
904 NOTE: Use this macro with caution in multibyte buffers (it is not
905 worth using this macro in unibyte buffers of course). Use of
906 `(set-buffer-multibyte t)', which is run finally, is generally
907 harmful since it is likely to modify existing data in the buffer.
908 For instance, it converts \"\\300\\255\" into \"\\255\" in
910 (let ((multibyte (make-symbol "multibyte"))
911 (buffer (make-symbol "buffer")))
913 (let ((,multibyte enable-multibyte-characters)
914 (,buffer (current-buffer)))
916 (let (default-enable-multibyte-characters)
917 (set-buffer-multibyte nil)
920 (set-buffer-multibyte ,multibyte)))
921 (let (default-enable-multibyte-characters)
923 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
924 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
926 (defmacro mm-with-unibyte (&rest forms)
927 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
928 `(let (default-enable-multibyte-characters)
930 (put 'mm-with-unibyte 'lisp-indent-function 0)
931 (put 'mm-with-unibyte 'edebug-form-spec '(body))
933 (defmacro mm-with-multibyte (&rest forms)
934 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
935 `(let ((default-enable-multibyte-characters t))
937 (put 'mm-with-multibyte 'lisp-indent-function 0)
938 (put 'mm-with-multibyte 'edebug-form-spec '(body))
940 (defun mm-find-charset-region (b e)
941 "Return a list of Emacs charsets in the region B to E."
943 ((and (mm-multibyte-p)
944 (fboundp 'find-charset-region))
945 ;; Remove composition since the base charsets have been included.
946 ;; Remove eight-bit-*, treat them as ascii.
947 (let ((css (find-charset-region b e)))
948 (mapcar (lambda (cs) (setq css (delq cs css)))
949 '(composition eight-bit-control eight-bit-graphic
953 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
956 (narrow-to-region b e)
957 (goto-char (point-min))
958 (skip-chars-forward "\0-\177")
963 (and (boundp 'current-language-environment)
964 (car (last (assq 'charset
965 (assoc current-language-environment
966 language-info-alist))))))
967 (if (eq charset 'ascii) (setq charset nil))
970 (car (last (assq mail-parse-charset
971 mm-mime-mule-charset-alist)))))
972 (list 'ascii (or charset 'latin-iso8859-1)))))))))
974 (defun mm-auto-mode-alist ()
975 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
976 (let ((alist auto-mode-alist)
979 (when (listp (cdar alist))
980 (push (car alist) out))
984 (defvar mm-inhibit-file-name-handlers
985 '(jka-compr-handler image-file-handler)
986 "A list of handlers doing (un)compression (etc) thingies.")
988 (defun mm-insert-file-contents (filename &optional visit beg end replace
990 "Like `insert-file-contents', but only reads in the file.
991 A buffer may be modified in several ways after reading into the buffer due
992 to advanced Emacs features, such as file-name-handlers, format decoding,
993 `find-file-hooks', etc.
994 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
995 This function ensures that none of these modifications will take place."
996 (let* ((format-alist nil)
997 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
998 (default-major-mode 'fundamental-mode)
999 (enable-local-variables nil)
1000 (after-insert-file-functions nil)
1001 (enable-local-eval nil)
1002 (inhibit-file-name-operation (if inhibit
1003 'insert-file-contents
1004 inhibit-file-name-operation))
1005 (inhibit-file-name-handlers
1007 (append mm-inhibit-file-name-handlers
1008 inhibit-file-name-handlers)
1009 inhibit-file-name-handlers))
1010 (ffh (if (boundp 'find-file-hook)
1013 (val (symbol-value ffh)))
1016 (insert-file-contents filename visit beg end replace)
1019 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1020 "Append the contents of the region to the end of file FILENAME.
1021 When called from a function, expects three arguments,
1022 START, END and FILENAME. START and END are buffer positions
1023 saying what text to write.
1024 Optional fourth argument specifies the coding system to use when
1026 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1027 (let ((coding-system-for-write
1028 (or codesys mm-text-coding-system-for-write
1029 mm-text-coding-system))
1030 (inhibit-file-name-operation (if inhibit
1032 inhibit-file-name-operation))
1033 (inhibit-file-name-handlers
1035 (append mm-inhibit-file-name-handlers
1036 inhibit-file-name-handlers)
1037 inhibit-file-name-handlers)))
1038 (write-region start end filename t 'no-message)
1039 (message "Appended to %s" filename)))
1041 (defun mm-write-region (start end filename &optional append visit lockname
1042 coding-system inhibit)
1044 "Like `write-region'.
1045 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1046 (let ((coding-system-for-write
1047 (or coding-system mm-text-coding-system-for-write
1048 mm-text-coding-system))
1049 (inhibit-file-name-operation (if inhibit
1051 inhibit-file-name-operation))
1052 (inhibit-file-name-handlers
1054 (append mm-inhibit-file-name-handlers
1055 inhibit-file-name-handlers)
1056 inhibit-file-name-handlers)))
1057 (write-region start end filename append visit lockname)))
1059 ;; It is not a MIME function, but some MIME functions use it.
1060 (if (and (fboundp 'make-temp-file)
1062 (let ((def (symbol-function 'make-temp-file)))
1063 (and (byte-code-function-p def)
1064 (setq def (if (fboundp 'compiled-function-arglist)
1066 (eval (list 'compiled-function-arglist def))
1069 (eq (nth 3 def) 'suffix)))))
1070 (defalias 'mm-make-temp-file 'make-temp-file)
1071 ;; Stolen (and modified for XEmacs) from Emacs 22.
1072 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1073 "Create a temporary file.
1074 The returned file name (created by appending some random characters at the end
1075 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1076 is guaranteed to point to a newly created empty file.
1077 You can then use `write-region' to write new data into the file.
1079 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1081 If SUFFIX is non-nil, add that at the end of the file name."
1082 (let ((umask (default-file-modes))
1086 ;; Create temp files with strict access rights. It's easy to
1087 ;; loosen them later, whereas it's impossible to close the
1088 ;; time-window of loose permissions otherwise.
1089 (set-default-file-modes 448)
1090 (while (condition-case err
1096 (if (fboundp 'temp-directory)
1099 temporary-file-directory))))
1101 (setq file (concat file suffix)))
1103 (make-directory file)
1104 (if (featurep 'xemacs)
1105 ;; NOTE: This is unsafe if XEmacs users
1106 ;; don't use a secure temp directory.
1107 (if (file-exists-p file)
1108 (signal 'file-already-exists
1109 (list "File exists" file))
1110 (write-region "" nil file nil 'silent))
1111 (write-region "" nil file nil 'silent
1114 (file-already-exists t)
1115 ;; The XEmacs version of `make-directory' issues
1117 (file-error (or (and (featurep 'xemacs)
1118 (file-exists-p file))
1119 (signal (car err) (cdr err)))))
1120 ;; the file was somehow created by someone else between
1121 ;; `make-temp-name' and `write-region', let's try again.
1125 (set-default-file-modes umask)))))
1127 (defun mm-image-load-path (&optional package)
1129 (dolist (path load-path (nreverse result))
1132 (setq dir (concat (file-name-directory
1133 (directory-file-name path))
1134 "etc/images/" (or package "gnus/")))))
1136 (push path result))))
1138 ;; Fixme: This doesn't look useful where it's used.
1139 (if (fboundp 'detect-coding-region)
1140 (defun mm-detect-coding-region (start end)
1141 "Like `detect-coding-region' except returning the best one."
1142 (let ((coding-systems
1143 (detect-coding-region start end)))
1144 (or (car-safe coding-systems)
1146 (defun mm-detect-coding-region (start end)
1147 (let ((point (point)))
1149 (skip-chars-forward "\0-\177" end)
1151 (if (eq (point) end) 'ascii (mm-guess-charset))
1152 (goto-char point)))))
1154 (if (fboundp 'coding-system-get)
1155 (defun mm-detect-mime-charset-region (start end)
1156 "Detect MIME charset of the text in the region between START and END."
1157 (let ((cs (mm-detect-coding-region start end)))
1158 (or (coding-system-get cs :mime-charset)
1159 (coding-system-get cs 'mime-charset))))
1160 (defun mm-detect-mime-charset-region (start end)
1161 "Detect MIME charset of the text in the region between START and END."
1162 (let ((cs (mm-detect-coding-region start end)))
1166 (unless (fboundp 'coding-system-to-mime-charset)
1167 (defalias 'coding-system-to-mime-charset 'ignore)))
1169 (defun mm-coding-system-to-mime-charset (coding-system)
1170 "Return the MIME charset corresponding to CODING-SYSTEM.
1171 To make this function work with XEmacs, the APEL package is required."
1173 (or (and (fboundp 'coding-system-get)
1174 (or (coding-system-get coding-system :mime-charset)
1175 (coding-system-get coding-system 'mime-charset)))
1176 (and (featurep 'xemacs)
1177 (or (and (fboundp 'coding-system-to-mime-charset)
1178 (not (eq (symbol-function 'coding-system-to-mime-charset)
1180 (and (condition-case nil
1183 (fboundp 'coding-system-to-mime-charset)))
1184 (coding-system-to-mime-charset coding-system)))))
1187 (require 'jka-compr))
1189 (defun mm-decompress-buffer (filename &optional inplace force)
1190 "Decompress buffer's contents, depending on jka-compr.
1191 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1192 agrees with `jka-compr-compression-info-list', decompression is done.
1193 Signal an error if FORCE is neither nil nor t and compressed data are
1194 not decompressed because `auto-compression-mode' is disabled.
1195 If INPLACE is nil, return decompressed data or nil without modifying
1196 the buffer. Otherwise, replace the buffer's contents with the
1197 decompressed data. The buffer's multibyteness must be turned off."
1200 (prog1 t (require 'jka-compr))
1201 (and (fboundp 'jka-compr-installed-p)
1202 (jka-compr-installed-p))))
1203 (let ((info (jka-compr-get-compression-info filename)))
1205 (unless (or (memq force (list nil t))
1206 (jka-compr-installed-p))
1208 (let ((prog (jka-compr-info-uncompress-program info))
1209 (args (jka-compr-info-uncompress-args info))
1210 (msg (format "%s %s..."
1211 (jka-compr-info-uncompress-message info)
1213 (err-file (jka-compr-make-temp-name))
1214 (cur (current-buffer))
1215 (coding-system-for-read mm-binary-coding-system)
1216 (coding-system-for-write mm-binary-coding-system)
1219 (mm-with-unibyte-buffer
1220 (insert-buffer-substring cur)
1223 (unless (memq (apply 'call-process-region
1224 (point-min) (point-max)
1225 prog t (list t err-file) nil args)
1226 jka-compr-acceptable-retval-list)
1230 (delete "" (split-string
1232 (insert-file-contents err-file)
1238 (format "Error while executing \"%s %s < %s\""
1239 prog (mapconcat 'identity args " ")
1241 (setq retval (buffer-string)))
1243 (setq err-msg (error-message-string err)))))
1244 (when (file-exists-p err-file)
1245 (ignore-errors (jka-compr-delete-temp-file err-file)))
1248 (delete-region (point-min) (point-max))
1251 (message "%s" (or err-msg (concat msg "done")))
1255 (unless (fboundp 'coding-system-name)
1256 (defalias 'coding-system-name 'ignore))
1257 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1258 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1259 (unless (fboundp 'find-operation-coding-system)
1260 (defalias 'find-operation-coding-system 'ignore)))
1262 (defun mm-find-buffer-file-coding-system (&optional filename)
1263 "Find coding system used to decode the contents of the current buffer.
1264 This function looks for the coding system magic cookie or examines the
1265 coding system specified by `file-coding-system-alist' being associated
1266 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1267 gzip, bzip2, etc. are allowed."
1269 (setq filename buffer-file-name))
1271 (let ((decomp (unless ;; No worth to examine charset of tar files.
1274 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1276 (mm-decompress-buffer filename nil t))))
1278 (set-buffer (let (default-enable-multibyte-characters)
1279 (generate-new-buffer " *temp*")))
1281 (setq filename (file-name-sans-extension filename)))
1282 (goto-char (point-min))
1285 ((boundp 'set-auto-coding-function) ;; Emacs
1287 (or (funcall (symbol-value 'set-auto-coding-function)
1288 filename (- (point-max) (point-min)))
1289 (car (find-operation-coding-system 'insert-file-contents
1291 (let (auto-coding-alist)
1293 (funcall (symbol-value 'set-auto-coding-function)
1294 nil (- (point-max) (point-min)))
1296 ((featurep 'file-coding) ;; XEmacs
1297 (let ((case-fold-search t)
1298 (end (point-at-eol))
1301 (and (re-search-forward "-\\*-+[\t ]*" end t)
1303 (setq start (match-end 0))
1304 (re-search-forward "[\t ]*-+\\*-" end t))
1306 (setq end (match-beginning 0))
1308 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1310 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1312 (find-coding-system (setq codesys
1313 (intern (match-string 1))))
1315 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1318 (setq start (match-end 0))
1319 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1321 (setq end (match-beginning 0))
1324 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1326 (find-coding-system (setq codesys
1327 (intern (match-string 1))))
1330 (goto-char (point-min))
1331 (setq case-fold-search nil)
1332 (re-search-forward "^;;;coding system: "
1333 ;;(+ (point-min) 3000) t))
1335 (looking-at "[^\t\n\r ]+")
1337 (setq codesys (intern (match-string 0))))
1341 (find-file-coding-system-for-read-from-filename
1343 (coding-system-name (coding-system-base codesys)))))))
1345 (kill-buffer (current-buffer)))))))
1349 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1350 ;;; mm-util.el ends here