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 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
365 (charsetp 'unicode-a)
366 (not (mm-coding-system-p 'mule-utf-8)))
367 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
368 ;; If we have utf-8 we're in Mule 5+.
371 (coding-system-get 'mule-utf-8 'safe-charsets)))))
372 "Alist of MIME-charset/MULE-charsets.")
374 (defun mm-enrich-utf-8-by-mule-ucs ()
375 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
376 This function will run when the `un-define' module is loaded under
377 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
378 with Mule charsets. It is completely useless for Emacs."
379 (unless (cdr (delete '(mm-enrich-utf-8-by-mule-ucs)
380 (assoc "un-define" after-load-alist)))
381 (setq after-load-alist
382 (delete '("un-define") after-load-alist)))
383 (when (boundp 'unicode-basic-translation-charset-order-list)
389 'unicode-basic-translation-charset-order-list))))
390 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
393 (setq mm-mime-mule-charset-alist
394 (nconc mm-mime-mule-charset-alist
395 (list (cons 'utf-8 val))))))
398 ;; Correct by construction, but should be unnecessary for Emacs:
399 (if (featurep 'xemacs)
400 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
401 (when (and (fboundp 'coding-system-list)
402 (fboundp 'sort-coding-systems))
403 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
407 mime (or (coding-system-get cs :mime-charset) ; Emacs 23 (unicode)
408 (coding-system-get cs 'mime-charset)))
410 (not (eq t (setq mule
411 (coding-system-get cs 'safe-charsets))))
412 (not (assq mime alist)))
413 (push (cons mime (delq 'ascii mule)) alist)))
414 (setq mm-mime-mule-charset-alist (nreverse alist)))))
416 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
417 "A list of special charsets.
418 Valid elements include:
419 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
420 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
423 (defvar mm-iso-8859-15-compatible
424 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
425 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
426 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
428 (defvar mm-iso-8859-x-to-15-table
429 (and (fboundp 'coding-system-p)
430 (mm-coding-system-p 'iso-8859-15)
433 (if (mm-coding-system-p (car cs))
434 (let ((c (string-to-char
435 (decode-coding-string "\341" (car cs)))))
436 (cons (char-charset c)
439 (decode-coding-string "\341" 'iso-8859-15)) c)
440 (string-to-list (decode-coding-string (car (cdr cs))
443 mm-iso-8859-15-compatible))
444 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
446 (defcustom mm-coding-system-priorities
447 (if (boundp 'current-language-environment)
448 (let ((lang (symbol-value 'current-language-environment)))
449 (cond ((string= lang "Japanese")
450 ;; Japanese users prefer iso-2022-jp to euc-japan or
451 ;; shift_jis, however iso-8859-1 should be used when
452 ;; there are only ASCII text and Latin-1 characters.
453 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
454 "Preferred coding systems for encoding outgoing messages.
456 More than one suitable coding system may be found for some text.
457 By default, the coding system with the highest priority is used
458 to encode outgoing messages (see `sort-coding-systems'). If this
459 variable is set, it overrides the default priority."
461 :type '(repeat (symbol :tag "Coding system"))
465 (defvar mm-use-find-coding-systems-region
466 (fboundp 'find-coding-systems-region)
467 "Use `find-coding-systems-region' to find proper coding systems.
469 Setting it to nil is useful on Emacsen supporting Unicode if sending
470 mail with multiple parts is preferred to sending a Unicode one.")
472 ;;; Internal variables:
476 (defun mm-mule-charset-to-mime-charset (charset)
477 "Return the MIME charset corresponding to the given Mule CHARSET."
478 (if (and (fboundp 'find-coding-systems-for-charsets)
479 (fboundp 'sort-coding-systems))
480 (let ((css (sort (sort-coding-systems
481 (find-coding-systems-for-charsets (list charset)))
482 'mm-sort-coding-systems-predicate))
484 (while (and (not mime)
486 (when (setq cs (pop css))
487 (setq mime (or (coding-system-get cs :mime-charset)
488 (coding-system-get cs 'mime-charset)))))
490 (let ((alist (mapcar (lambda (cs)
491 (assq cs mm-mime-mule-charset-alist))
492 (sort (mapcar 'car mm-mime-mule-charset-alist)
493 'mm-sort-coding-systems-predicate)))
496 (when (memq charset (cdar alist))
497 (setq out (caar alist)
502 (defun mm-charset-to-coding-system (charset &optional lbt
504 "Return coding-system corresponding to CHARSET.
505 CHARSET is a symbol naming a MIME charset.
506 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
507 used as the line break code type of the coding system.
509 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
510 map undesired charset names to their replacement. This should
511 only be used for decoding, not for encoding."
512 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
513 (when (stringp charset)
514 (setq charset (intern (downcase charset))))
516 (setq charset (intern (format "%s-%s" charset lbt))))
520 ;; Running in a non-MULE environment.
521 ((or (null (mm-get-coding-system-list))
522 (not (fboundp 'coding-system-get)))
524 ;; Check override list quite early. Should only used for decoding, not for
527 (let ((cs (cdr (assq charset mm-charset-override-alist))))
528 (and cs (mm-coding-system-p cs) cs))))
530 ((eq charset 'us-ascii)
532 ;; Check to see whether we can handle this charset. (This depends
533 ;; on there being some coding system matching each `mime-charset'
534 ;; property defined, as there should be.)
535 ((and (mm-coding-system-p charset)
536 ;;; Doing this would potentially weed out incorrect charsets.
538 ;;; (eq charset (coding-system-get charset 'mime-charset))
541 ;; Eval expressions from `mm-charset-eval-alist'
542 ((let* ((el (assq charset mm-charset-eval-alist))
549 (condition-case nil (eval form) (error nil))
550 ;; (message "Failed to eval `%s'" form))
551 (mm-coding-system-p cs)
552 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
554 ;; Translate invalid charsets.
555 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
557 (mm-coding-system-p cs)
559 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
562 ;; Last resort: search the coding system list for entries which
563 ;; have the right mime-charset in case the canonical name isn't
564 ;; defined (though it should be).
566 ;; mm-get-coding-system-list returns a list of cs without lbt.
568 (dolist (c (mm-get-coding-system-list))
570 (eq charset (or (coding-system-get c :mime-charset)
571 (coding-system-get c 'mime-charset))))
574 ;; Warn the user about unknown charset:
575 (if (fboundp 'gnus-message)
576 (gnus-message 7 "Unknown charset: %s" charset)
577 (message "Unknown charset: %s" charset)))
581 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
582 (boundp 'default-enable-multibyte-characters)
583 default-enable-multibyte-characters
584 (fboundp 'set-buffer-multibyte))
585 "True in Emacs with Mule.")
588 (defun mm-enable-multibyte ()
589 "Set the multibyte flag of the current buffer.
590 Only do this if the default value of `enable-multibyte-characters' is
591 non-nil. This is a no-op in XEmacs."
592 (set-buffer-multibyte 'to))
593 (defalias 'mm-enable-multibyte 'ignore))
596 (defun mm-disable-multibyte ()
597 "Unset the multibyte flag of in the current buffer.
598 This is a no-op in XEmacs."
599 (set-buffer-multibyte nil))
600 (defalias 'mm-disable-multibyte 'ignore)))
602 (defun mm-preferred-coding-system (charset)
603 ;; A typo in some Emacs versions.
604 (or (get-charset-property charset 'preferred-coding-system)
605 (get-charset-property charset 'prefered-coding-system)))
607 ;; Mule charsets shouldn't be used.
608 (defsubst mm-guess-charset ()
609 "Guess Mule charset from the language environment."
611 mail-parse-mule-charset ;; cached mule-charset
613 (setq mail-parse-mule-charset
614 (and (boundp 'current-language-environment)
617 (assoc current-language-environment
618 language-info-alist))))))
619 (if (or (not mail-parse-mule-charset)
620 (eq mail-parse-mule-charset 'ascii))
621 (setq mail-parse-mule-charset
622 (or (car (last (assq mail-parse-charset
623 mm-mime-mule-charset-alist)))
626 mail-parse-mule-charset)))
628 (defun mm-charset-after (&optional pos)
629 "Return charset of a character in current buffer at position POS.
630 If POS is nil, it defauls to the current point.
631 If POS is out of range, the value is nil.
632 If the charset is `composition', return the actual one."
633 (let ((char (char-after pos)) charset)
634 (if (< (mm-char-int char) 128)
635 (setq charset 'ascii)
636 ;; charset-after is fake in some Emacsen.
637 (setq charset (and (fboundp 'char-charset) (char-charset char)))
638 (if (eq charset 'composition) ; Mule 4
639 (let ((p (or pos (point))))
640 (cadr (find-charset-region p (1+ p))))
641 (if (and charset (not (memq charset '(ascii eight-bit-control
642 eight-bit-graphic))))
644 (mm-guess-charset))))))
646 (defun mm-mime-charset (charset)
647 "Return the MIME charset corresponding to the given Mule CHARSET."
648 (if (eq charset 'unknown)
649 (error "The message contains non-printable characters, please use attachment"))
650 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
651 ;; This exists in Emacs 20.
653 (and (mm-preferred-coding-system charset)
654 (or (coding-system-get
655 (mm-preferred-coding-system charset) :mime-charset)
657 (mm-preferred-coding-system charset) 'mime-charset)))
658 (and (eq charset 'ascii)
660 (mm-preferred-coding-system charset)
661 (mm-mule-charset-to-mime-charset charset))
662 ;; This is for XEmacs.
663 (mm-mule-charset-to-mime-charset charset)))
665 (if (fboundp 'delete-dups)
666 (defalias 'mm-delete-duplicates 'delete-dups)
667 (defun mm-delete-duplicates (list)
668 "Destructively remove `equal' duplicates from LIST.
669 Store the result in LIST and return it. LIST must be a proper list.
670 Of several `equal' occurrences of an element in LIST, the first
673 This is a compatibility function for Emacsen without `delete-dups'."
674 ;; Code from `subr.el' in Emacs 22:
677 (setcdr tail (delete (car tail) (cdr tail)))
678 (setq tail (cdr tail))))
681 ;; Fixme: This is used in places when it should be testing the
682 ;; default multibyteness. See mm-default-multibyte-p.
684 (if (and (not (featurep 'xemacs))
685 (boundp 'enable-multibyte-characters))
686 (defun mm-multibyte-p ()
687 "Non-nil if multibyte is enabled in the current buffer."
688 enable-multibyte-characters)
689 (defun mm-multibyte-p () (featurep 'mule))))
691 (defun mm-default-multibyte-p ()
692 "Return non-nil if the session is multibyte.
693 This affects whether coding conversion should be attempted generally."
695 (if (boundp 'default-enable-multibyte-characters)
696 default-enable-multibyte-characters
699 (defun mm-iso-8859-x-to-15-region (&optional b e)
700 (if (fboundp 'char-charset)
701 (let (charset item c inconvertible)
703 (if e (narrow-to-region b e))
704 (goto-char (point-min))
705 (skip-chars-forward "\0-\177")
708 ((not (setq item (assq (char-charset (setq c (char-after)))
709 mm-iso-8859-x-to-15-table)))
711 ((memq c (cdr (cdr item)))
712 (setq inconvertible t)
715 (insert-before-markers (prog1 (+ c (car (cdr item)))
717 (skip-chars-forward "\0-\177")))
718 (not inconvertible))))
720 (defun mm-sort-coding-systems-predicate (a b)
723 ;; Note: invalid entries are dropped silently
724 (and (setq cs (mm-coding-system-p cs))
725 (coding-system-base cs)))
726 mm-coding-system-priorities)))
727 (and (setq a (mm-coding-system-p a))
728 (if (setq b (mm-coding-system-p b))
729 (> (length (memq (coding-system-base a) priorities))
730 (length (memq (coding-system-base b) priorities)))
734 (autoload 'latin-unity-massage-name "latin-unity")
735 (autoload 'latin-unity-maybe-remap "latin-unity")
736 (autoload 'latin-unity-representations-feasible-region "latin-unity")
737 (autoload 'latin-unity-representations-present-region "latin-unity")
738 (defvar latin-unity-coding-systems)
739 (defvar latin-unity-ucs-list))
741 (defun mm-xemacs-find-mime-charset-1 (begin end)
742 "Determine which MIME charset to use to send region as message.
743 This uses the XEmacs-specific latin-unity package to better handle the
744 case where identical characters from diverse ISO-8859-? character sets
745 can be encoded using a single one of the corresponding coding systems.
747 It treats `mm-coding-system-priorities' as the list of preferred
748 coding systems; a useful example setting for this list in Western
749 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
750 to the very standard Latin 1 coding system, and only move to coding
751 systems that are less supported as is necessary to encode the
752 characters that exist in the buffer.
754 Latin Unity doesn't know about those non-ASCII Roman characters that
755 are available in various East Asian character sets. As such, its
756 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
757 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
758 But this is very much a corner case, so don't worry about it."
759 (let ((systems mm-coding-system-priorities) csets psets curset)
761 ;; Load the Latin Unity library, if available.
762 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
763 (require 'latin-unity))
765 ;; Now, can we use it?
766 (if (featurep 'latin-unity)
768 (setq csets (latin-unity-representations-feasible-region begin end)
769 psets (latin-unity-representations-present-region begin end))
773 ;; Pass back the first coding system in the preferred list
774 ;; that can encode the whole region.
775 (dolist (curset systems)
776 (setq curset (latin-unity-massage-name 'buffer-default curset))
778 ;; If the coding system is a universal coding system, then
779 ;; it can certainly encode all the characters in the region.
780 (if (memq curset latin-unity-ucs-list)
781 (throw 'done (list curset)))
783 ;; If a coding system isn't universal, and isn't in
784 ;; the list that latin unity knows about, we can't
785 ;; decide whether to use it here. Leave that until later
786 ;; in `mm-find-mime-charset-region' function, whence we
788 (unless (memq curset latin-unity-coding-systems)
791 ;; Right, we know about this coding system, and it may
792 ;; conceivably be able to encode all the characters in
794 (if (latin-unity-maybe-remap begin end curset csets psets t)
795 (throw 'done (list curset))))
797 ;; Can't encode using anything from the
798 ;; `mm-coding-system-priorities' list.
799 ;; Leave `mm-find-mime-charset' to do most of the work.
802 ;; Right, latin unity isn't available; let `mm-find-charset-region'
803 ;; take its default action, which equally applies to GNU Emacs.
806 (defmacro mm-xemacs-find-mime-charset (begin end)
807 (when (featurep 'xemacs)
808 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
810 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
811 "Return the MIME charsets needed to encode the region between B and E.
812 nil means ASCII, a single-element list represents an appropriate MIME
813 charset, and a longer list means no appropriate charset."
815 ;; The return possibilities of this function are a mess...
816 (or (and (mm-multibyte-p)
817 mm-use-find-coding-systems-region
818 ;; Find the mime-charset of the most preferred coding
819 ;; system that has one.
820 (let ((systems (find-coding-systems-region b e)))
821 (when mm-coding-system-priorities
823 (sort systems 'mm-sort-coding-systems-predicate)))
824 (setq systems (delq 'compound-text systems))
825 (unless (equal systems '(undecided))
827 (let* ((head (pop systems))
828 (cs (or (coding-system-get head :mime-charset)
829 (coding-system-get head 'mime-charset))))
830 ;; The mime-charset (`x-ctext') of
831 ;; `compound-text' is not in the IANA list. We
832 ;; shouldn't normally use anything here with a
833 ;; mime-charset having an `x-' prefix.
834 ;; Fixme: Allow this to be overridden, since
835 ;; there is existing use of x-ctext.
836 ;; Also people apparently need the coding system
837 ;; `iso-2022-jp-3' (which Mule-UCS defines with
838 ;; mime-charset, though it's not valid).
840 (not (string-match "^[Xx]-" (symbol-name cs)))
841 ;; UTF-16 of any variety is invalid for
842 ;; text parts and, unfortunately, has
843 ;; mime-charset defined both in Mule-UCS
844 ;; and versions of Emacs. (The name
845 ;; might be `mule-utf-16...' or
847 (not (string-match "utf-16" (symbol-name cs))))
849 charsets (list cs))))))
851 ;; If we're XEmacs, and some coding system is appropriate,
852 ;; mm-xemacs-find-mime-charset will return an appropriate list.
853 ;; Otherwise, we'll get nil, and the next setq will get invoked.
854 (setq charsets (mm-xemacs-find-mime-charset b e))
856 ;; We're not multibyte, or a single coding system won't cover it.
858 (mm-delete-duplicates
859 (mapcar 'mm-mime-charset
861 (mm-find-charset-region b e))))))
862 (if (and (> (length charsets) 1)
863 (memq 'iso-8859-15 charsets)
864 (memq 'iso-8859-15 hack-charsets)
865 (save-excursion (mm-iso-8859-x-to-15-region b e)))
866 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
867 mm-iso-8859-15-compatible))
868 (if (and (memq 'iso-2022-jp-2 charsets)
869 (memq 'iso-2022-jp-2 hack-charsets))
870 (setq charsets (delq 'iso-2022-jp charsets)))
871 ;; Attempt to reduce the number of charsets if utf-8 is available.
872 (if (and (featurep 'xemacs)
873 (> (length charsets) 1)
874 (mm-coding-system-p 'utf-8))
875 (let ((mm-coding-system-priorities
876 (cons 'utf-8 mm-coding-system-priorities)))
878 (mm-delete-duplicates
879 (mapcar 'mm-mime-charset
881 (mm-find-charset-region b e)))))))
884 (defmacro mm-with-unibyte-buffer (&rest forms)
885 "Create a temporary buffer, and evaluate FORMS there like `progn'.
886 Use unibyte mode for this."
887 `(let (default-enable-multibyte-characters)
888 (with-temp-buffer ,@forms)))
889 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
890 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
892 (defmacro mm-with-multibyte-buffer (&rest forms)
893 "Create a temporary buffer, and evaluate FORMS there like `progn'.
894 Use multibyte mode for this."
895 `(let ((default-enable-multibyte-characters t))
896 (with-temp-buffer ,@forms)))
897 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
898 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
900 (defmacro mm-with-unibyte-current-buffer (&rest forms)
901 "Evaluate FORMS with current buffer temporarily made unibyte.
902 Also bind `default-enable-multibyte-characters' to nil.
903 Equivalent to `progn' in XEmacs
905 NOTE: Use this macro with caution in multibyte buffers (it is not
906 worth using this macro in unibyte buffers of course). Use of
907 `(set-buffer-multibyte t)', which is run finally, is generally
908 harmful since it is likely to modify existing data in the buffer.
909 For instance, it converts \"\\300\\255\" into \"\\255\" in
911 (let ((multibyte (make-symbol "multibyte"))
912 (buffer (make-symbol "buffer")))
914 (let ((,multibyte enable-multibyte-characters)
915 (,buffer (current-buffer)))
917 (let (default-enable-multibyte-characters)
918 (set-buffer-multibyte nil)
921 (set-buffer-multibyte ,multibyte)))
922 (let (default-enable-multibyte-characters)
924 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
925 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
927 (defmacro mm-with-unibyte (&rest forms)
928 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
929 `(let (default-enable-multibyte-characters)
931 (put 'mm-with-unibyte 'lisp-indent-function 0)
932 (put 'mm-with-unibyte 'edebug-form-spec '(body))
934 (defmacro mm-with-multibyte (&rest forms)
935 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
936 `(let ((default-enable-multibyte-characters t))
938 (put 'mm-with-multibyte 'lisp-indent-function 0)
939 (put 'mm-with-multibyte 'edebug-form-spec '(body))
941 (defun mm-find-charset-region (b e)
942 "Return a list of Emacs charsets in the region B to E."
944 ((and (mm-multibyte-p)
945 (fboundp 'find-charset-region))
946 ;; Remove composition since the base charsets have been included.
947 ;; Remove eight-bit-*, treat them as ascii.
948 (let ((css (find-charset-region b e)))
949 (mapcar (lambda (cs) (setq css (delq cs css)))
950 '(composition eight-bit-control eight-bit-graphic
954 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
957 (narrow-to-region b e)
958 (goto-char (point-min))
959 (skip-chars-forward "\0-\177")
964 (and (boundp 'current-language-environment)
965 (car (last (assq 'charset
966 (assoc current-language-environment
967 language-info-alist))))))
968 (if (eq charset 'ascii) (setq charset nil))
971 (car (last (assq mail-parse-charset
972 mm-mime-mule-charset-alist)))))
973 (list 'ascii (or charset 'latin-iso8859-1)))))))))
975 (defun mm-auto-mode-alist ()
976 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
977 (let ((alist auto-mode-alist)
980 (when (listp (cdar alist))
981 (push (car alist) out))
985 (defvar mm-inhibit-file-name-handlers
986 '(jka-compr-handler image-file-handler)
987 "A list of handlers doing (un)compression (etc) thingies.")
989 (defun mm-insert-file-contents (filename &optional visit beg end replace
991 "Like `insert-file-contents', but only reads in the file.
992 A buffer may be modified in several ways after reading into the buffer due
993 to advanced Emacs features, such as file-name-handlers, format decoding,
994 `find-file-hooks', etc.
995 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
996 This function ensures that none of these modifications will take place."
997 (let* ((format-alist nil)
998 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
999 (default-major-mode 'fundamental-mode)
1000 (enable-local-variables nil)
1001 (after-insert-file-functions nil)
1002 (enable-local-eval nil)
1003 (inhibit-file-name-operation (if inhibit
1004 'insert-file-contents
1005 inhibit-file-name-operation))
1006 (inhibit-file-name-handlers
1008 (append mm-inhibit-file-name-handlers
1009 inhibit-file-name-handlers)
1010 inhibit-file-name-handlers))
1011 (ffh (if (boundp 'find-file-hook)
1014 (val (symbol-value ffh)))
1017 (insert-file-contents filename visit beg end replace)
1020 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1021 "Append the contents of the region to the end of file FILENAME.
1022 When called from a function, expects three arguments,
1023 START, END and FILENAME. START and END are buffer positions
1024 saying what text to write.
1025 Optional fourth argument specifies the coding system to use when
1027 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1028 (let ((coding-system-for-write
1029 (or codesys mm-text-coding-system-for-write
1030 mm-text-coding-system))
1031 (inhibit-file-name-operation (if inhibit
1033 inhibit-file-name-operation))
1034 (inhibit-file-name-handlers
1036 (append mm-inhibit-file-name-handlers
1037 inhibit-file-name-handlers)
1038 inhibit-file-name-handlers)))
1039 (write-region start end filename t 'no-message)
1040 (message "Appended to %s" filename)))
1042 (defun mm-write-region (start end filename &optional append visit lockname
1043 coding-system inhibit)
1045 "Like `write-region'.
1046 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1047 (let ((coding-system-for-write
1048 (or coding-system mm-text-coding-system-for-write
1049 mm-text-coding-system))
1050 (inhibit-file-name-operation (if inhibit
1052 inhibit-file-name-operation))
1053 (inhibit-file-name-handlers
1055 (append mm-inhibit-file-name-handlers
1056 inhibit-file-name-handlers)
1057 inhibit-file-name-handlers)))
1058 (write-region start end filename append visit lockname)))
1060 ;; It is not a MIME function, but some MIME functions use it.
1061 (if (and (fboundp 'make-temp-file)
1063 (let ((def (symbol-function 'make-temp-file)))
1064 (and (byte-code-function-p def)
1065 (setq def (if (fboundp 'compiled-function-arglist)
1067 (eval (list 'compiled-function-arglist def))
1070 (eq (nth 3 def) 'suffix)))))
1071 (defalias 'mm-make-temp-file 'make-temp-file)
1072 ;; Stolen (and modified for XEmacs) from Emacs 22.
1073 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1074 "Create a temporary file.
1075 The returned file name (created by appending some random characters at the end
1076 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1077 is guaranteed to point to a newly created empty file.
1078 You can then use `write-region' to write new data into the file.
1080 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1082 If SUFFIX is non-nil, add that at the end of the file name."
1083 (let ((umask (default-file-modes))
1087 ;; Create temp files with strict access rights. It's easy to
1088 ;; loosen them later, whereas it's impossible to close the
1089 ;; time-window of loose permissions otherwise.
1090 (set-default-file-modes 448)
1091 (while (condition-case err
1097 (if (fboundp 'temp-directory)
1100 temporary-file-directory))))
1102 (setq file (concat file suffix)))
1104 (make-directory file)
1105 (if (featurep 'xemacs)
1106 ;; NOTE: This is unsafe if XEmacs users
1107 ;; don't use a secure temp directory.
1108 (if (file-exists-p file)
1109 (signal 'file-already-exists
1110 (list "File exists" file))
1111 (write-region "" nil file nil 'silent))
1112 (write-region "" nil file nil 'silent
1115 (file-already-exists t)
1116 ;; The XEmacs version of `make-directory' issues
1118 (file-error (or (and (featurep 'xemacs)
1119 (file-exists-p file))
1120 (signal (car err) (cdr err)))))
1121 ;; the file was somehow created by someone else between
1122 ;; `make-temp-name' and `write-region', let's try again.
1126 (set-default-file-modes umask)))))
1128 (defun mm-image-load-path (&optional package)
1130 (dolist (path load-path (nreverse result))
1133 (setq dir (concat (file-name-directory
1134 (directory-file-name path))
1135 "etc/images/" (or package "gnus/")))))
1137 (push path result))))
1139 ;; Fixme: This doesn't look useful where it's used.
1140 (if (fboundp 'detect-coding-region)
1141 (defun mm-detect-coding-region (start end)
1142 "Like `detect-coding-region' except returning the best one."
1143 (let ((coding-systems
1144 (detect-coding-region start end)))
1145 (or (car-safe coding-systems)
1147 (defun mm-detect-coding-region (start end)
1148 (let ((point (point)))
1150 (skip-chars-forward "\0-\177" end)
1152 (if (eq (point) end) 'ascii (mm-guess-charset))
1153 (goto-char point)))))
1155 (if (fboundp 'coding-system-get)
1156 (defun mm-detect-mime-charset-region (start end)
1157 "Detect MIME charset of the text in the region between START and END."
1158 (let ((cs (mm-detect-coding-region start end)))
1159 (or (coding-system-get cs :mime-charset)
1160 (coding-system-get cs 'mime-charset))))
1161 (defun mm-detect-mime-charset-region (start end)
1162 "Detect MIME charset of the text in the region between START and END."
1163 (let ((cs (mm-detect-coding-region start end)))
1167 (unless (fboundp 'coding-system-to-mime-charset)
1168 (defalias 'coding-system-to-mime-charset 'ignore)))
1170 (defun mm-coding-system-to-mime-charset (coding-system)
1171 "Return the MIME charset corresponding to CODING-SYSTEM.
1172 To make this function work with XEmacs, the APEL package is required."
1174 (or (and (fboundp 'coding-system-get)
1175 (or (coding-system-get coding-system :mime-charset)
1176 (coding-system-get coding-system 'mime-charset)))
1177 (and (featurep 'xemacs)
1178 (or (and (fboundp 'coding-system-to-mime-charset)
1179 (not (eq (symbol-function 'coding-system-to-mime-charset)
1181 (and (condition-case nil
1184 (fboundp 'coding-system-to-mime-charset)))
1185 (coding-system-to-mime-charset coding-system)))))
1188 (require 'jka-compr))
1190 (defun mm-decompress-buffer (filename &optional inplace force)
1191 "Decompress buffer's contents, depending on jka-compr.
1192 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1193 agrees with `jka-compr-compression-info-list', decompression is done.
1194 Signal an error if FORCE is neither nil nor t and compressed data are
1195 not decompressed because `auto-compression-mode' is disabled.
1196 If INPLACE is nil, return decompressed data or nil without modifying
1197 the buffer. Otherwise, replace the buffer's contents with the
1198 decompressed data. The buffer's multibyteness must be turned off."
1201 (prog1 t (require 'jka-compr))
1202 (and (fboundp 'jka-compr-installed-p)
1203 (jka-compr-installed-p))))
1204 (let ((info (jka-compr-get-compression-info filename)))
1206 (unless (or (memq force (list nil t))
1207 (jka-compr-installed-p))
1209 (let ((prog (jka-compr-info-uncompress-program info))
1210 (args (jka-compr-info-uncompress-args info))
1211 (msg (format "%s %s..."
1212 (jka-compr-info-uncompress-message info)
1214 (err-file (jka-compr-make-temp-name))
1215 (cur (current-buffer))
1216 (coding-system-for-read mm-binary-coding-system)
1217 (coding-system-for-write mm-binary-coding-system)
1220 (mm-with-unibyte-buffer
1221 (insert-buffer-substring cur)
1224 (unless (memq (apply 'call-process-region
1225 (point-min) (point-max)
1226 prog t (list t err-file) nil args)
1227 jka-compr-acceptable-retval-list)
1231 (delete "" (split-string
1233 (insert-file-contents err-file)
1239 (format "Error while executing \"%s %s < %s\""
1240 prog (mapconcat 'identity args " ")
1242 (setq retval (buffer-string)))
1244 (setq err-msg (error-message-string err)))))
1245 (when (file-exists-p err-file)
1246 (ignore-errors (jka-compr-delete-temp-file err-file)))
1249 (delete-region (point-min) (point-max))
1252 (message "%s" (or err-msg (concat msg "done")))
1256 (unless (fboundp 'coding-system-name)
1257 (defalias 'coding-system-name 'ignore))
1258 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1259 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1260 (unless (fboundp 'find-operation-coding-system)
1261 (defalias 'find-operation-coding-system 'ignore)))
1263 (defun mm-find-buffer-file-coding-system (&optional filename)
1264 "Find coding system used to decode the contents of the current buffer.
1265 This function looks for the coding system magic cookie or examines the
1266 coding system specified by `file-coding-system-alist' being associated
1267 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1268 gzip, bzip2, etc. are allowed."
1270 (setq filename buffer-file-name))
1272 (let ((decomp (unless ;; No worth to examine charset of tar files.
1275 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1277 (mm-decompress-buffer filename nil t))))
1279 (set-buffer (let (default-enable-multibyte-characters)
1280 (generate-new-buffer " *temp*")))
1282 (setq filename (file-name-sans-extension filename)))
1283 (goto-char (point-min))
1286 ((boundp 'set-auto-coding-function) ;; Emacs
1288 (or (funcall (symbol-value 'set-auto-coding-function)
1289 filename (- (point-max) (point-min)))
1290 (car (find-operation-coding-system 'insert-file-contents
1292 (let (auto-coding-alist)
1294 (funcall (symbol-value 'set-auto-coding-function)
1295 nil (- (point-max) (point-min)))
1297 ((featurep 'file-coding) ;; XEmacs
1298 (let ((case-fold-search t)
1299 (end (point-at-eol))
1302 (and (re-search-forward "-\\*-+[\t ]*" end t)
1304 (setq start (match-end 0))
1305 (re-search-forward "[\t ]*-+\\*-" end t))
1307 (setq end (match-beginning 0))
1309 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1311 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1313 (find-coding-system (setq codesys
1314 (intern (match-string 1))))
1316 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1319 (setq start (match-end 0))
1320 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1322 (setq end (match-beginning 0))
1325 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1327 (find-coding-system (setq codesys
1328 (intern (match-string 1))))
1331 (goto-char (point-min))
1332 (setq case-fold-search nil)
1333 (re-search-forward "^;;;coding system: "
1334 ;;(+ (point-min) 3000) t))
1336 (looking-at "[^\t\n\r ]+")
1338 (setq codesys (intern (match-string 0))))
1342 (find-file-coding-system-for-read-from-filename
1344 (coding-system-name (coding-system-base codesys)))))))
1346 (kill-buffer (current-buffer)))))))
1350 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1351 ;;; mm-util.el ends here