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))))
232 ;; Windows-31J is Windows Codepage 932.
233 ,@(if (and (not (mm-coding-system-p 'windows-31j))
234 (mm-coding-system-p 'cp932))
235 '((windows-31j . cp932)))
237 "A mapping from unknown or invalid charset names to the real charset names.")
239 (defcustom mm-charset-override-alist
240 `((iso-8859-1 . windows-1252))
241 "A mapping from undesired charset names to their replacement.
243 You may add pairs like (iso-8859-1 . windows-1252) here,
244 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
245 superset of iso-8859-1."
246 :type '(list (set :inline t
247 (const (iso-8859-1 . windows-1252))
248 (const (undecided . windows-1252)))
251 (cons (symbol :tag "From charset")
252 (symbol :tag "To charset"))))
253 :version "22.1" ;; Gnus 5.10.9
256 (defcustom mm-charset-eval-alist
257 (if (featurep 'xemacs)
258 nil ;; I don't know what would be useful for XEmacs.
259 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
260 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
261 (windows-1250 . (mm-codepage-setup 1250 t))
262 (windows-1251 . (mm-codepage-setup 1251 t))
263 (windows-1253 . (mm-codepage-setup 1253 t))
264 (windows-1257 . (mm-codepage-setup 1257 t))))
265 "An alist of (CHARSET . FORM) pairs.
266 If an article is encoded in an unknown CHARSET, FORM is
267 evaluated. This allows to load additional libraries providing
268 charsets on demand. If supported by your Emacs version, you
269 could use `autoload-coding-system' here."
270 :version "22.1" ;; Gnus 5.10.9
271 :type '(list (set :inline t
272 (const (windows-1250 . (mm-codepage-setup 1250 t)))
273 (const (windows-1251 . (mm-codepage-setup 1251 t)))
274 (const (windows-1253 . (mm-codepage-setup 1253 t)))
275 (const (windows-1257 . (mm-codepage-setup 1257 t)))
276 (const (cp850 . (mm-codepage-setup 850 nil))))
279 (cons (symbol :tag "charset")
280 (symbol :tag "form"))))
283 (defvar mm-binary-coding-system
285 ((mm-coding-system-p 'binary) 'binary)
286 ((mm-coding-system-p 'no-conversion) 'no-conversion)
288 "100% binary coding system.")
290 (defvar mm-text-coding-system
291 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
292 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
293 (and (mm-coding-system-p 'raw-text) 'raw-text))
294 mm-binary-coding-system)
295 "Text-safe coding system (For removing ^M).")
297 (defvar mm-text-coding-system-for-write nil
298 "Text coding system for write.")
300 (defvar mm-auto-save-coding-system
302 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
303 (if (memq system-type '(windows-nt ms-dos ms-windows))
304 (if (mm-coding-system-p 'utf-8-emacs-dos)
305 'utf-8-emacs-dos mm-binary-coding-system)
307 ((mm-coding-system-p 'emacs-mule)
308 (if (memq system-type '(windows-nt ms-dos ms-windows))
309 (if (mm-coding-system-p 'emacs-mule-dos)
310 'emacs-mule-dos mm-binary-coding-system)
312 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
313 (t mm-binary-coding-system))
314 "Coding system of auto save file.")
316 (defvar mm-universal-coding-system mm-auto-save-coding-system
317 "The universal coding system.")
319 ;; Fixme: some of the cars here aren't valid MIME charsets. That
320 ;; should only matter with XEmacs, though.
321 (defvar mm-mime-mule-charset-alist
323 (iso-8859-1 latin-iso8859-1)
324 (iso-8859-2 latin-iso8859-2)
325 (iso-8859-3 latin-iso8859-3)
326 (iso-8859-4 latin-iso8859-4)
327 (iso-8859-5 cyrillic-iso8859-5)
328 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
329 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
330 ;; charset is koi8-r, not iso-8859-5.
331 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
332 (iso-8859-6 arabic-iso8859-6)
333 (iso-8859-7 greek-iso8859-7)
334 (iso-8859-8 hebrew-iso8859-8)
335 (iso-8859-9 latin-iso8859-9)
336 (iso-8859-14 latin-iso8859-14)
337 (iso-8859-15 latin-iso8859-15)
338 (viscii vietnamese-viscii-lower)
339 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
340 (euc-kr korean-ksc5601)
341 (gb2312 chinese-gb2312)
342 (big5 chinese-big5-1 chinese-big5-2)
344 (thai-tis620 thai-tis620)
345 (windows-1251 cyrillic-iso8859-5)
346 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
347 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
348 latin-jisx0201 japanese-jisx0208-1978
349 chinese-gb2312 japanese-jisx0208
350 korean-ksc5601 japanese-jisx0212)
351 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
352 latin-jisx0201 japanese-jisx0208-1978
353 chinese-gb2312 japanese-jisx0208
354 korean-ksc5601 japanese-jisx0212
355 chinese-cns11643-1 chinese-cns11643-2)
356 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
357 cyrillic-iso8859-5 greek-iso8859-7
358 latin-jisx0201 japanese-jisx0208-1978
359 chinese-gb2312 japanese-jisx0208
360 korean-ksc5601 japanese-jisx0212
361 chinese-cns11643-1 chinese-cns11643-2
362 chinese-cns11643-3 chinese-cns11643-4
363 chinese-cns11643-5 chinese-cns11643-6
365 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
366 japanese-jisx0213-1 japanese-jisx0213-2)
367 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
368 ,(cond ((fboundp 'unicode-precedence-list)
369 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
370 (unicode-precedence-list)))))
371 ((or (not (fboundp 'charsetp)) ;; non-Mule case
372 (charsetp 'unicode-a)
373 (not (mm-coding-system-p 'mule-utf-8)))
374 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
375 (t ;; If we have utf-8 we're in Mule 5+.
378 (coding-system-get 'mule-utf-8 'safe-charsets))))))
379 "Alist of MIME-charset/MULE-charsets.")
381 (defun mm-enrich-utf-8-by-mule-ucs ()
382 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
383 This function will run when the `un-define' module is loaded under
384 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
385 with Mule charsets. It is completely useless for Emacs."
386 (when (boundp 'unicode-basic-translation-charset-order-list)
392 'unicode-basic-translation-charset-order-list))))
393 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
396 (setq mm-mime-mule-charset-alist
397 (nconc mm-mime-mule-charset-alist
398 (list (cons 'utf-8 val))))))
401 ;; Correct by construction, but should be unnecessary for Emacs:
402 (if (featurep 'xemacs)
403 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
404 (when (and (fboundp 'coding-system-list)
405 (fboundp 'sort-coding-systems))
406 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
410 mime (or (coding-system-get cs :mime-charset) ; Emacs 23 (unicode)
411 (coding-system-get cs 'mime-charset)))
413 (not (eq t (setq mule
414 (coding-system-get cs 'safe-charsets))))
415 (not (assq mime alist)))
416 (push (cons mime (delq 'ascii mule)) alist)))
417 (setq mm-mime-mule-charset-alist (nreverse alist)))))
419 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
420 "A list of special charsets.
421 Valid elements include:
422 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
423 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
426 (defvar mm-iso-8859-15-compatible
427 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
428 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
429 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
431 (defvar mm-iso-8859-x-to-15-table
432 (and (fboundp 'coding-system-p)
433 (mm-coding-system-p 'iso-8859-15)
436 (if (mm-coding-system-p (car cs))
437 (let ((c (string-to-char
438 (decode-coding-string "\341" (car cs)))))
439 (cons (char-charset c)
442 (decode-coding-string "\341" 'iso-8859-15)) c)
443 (string-to-list (decode-coding-string (car (cdr cs))
446 mm-iso-8859-15-compatible))
447 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
449 (defcustom mm-coding-system-priorities
450 (if (boundp 'current-language-environment)
451 (let ((lang (symbol-value 'current-language-environment)))
452 (cond ((string= lang "Japanese")
453 ;; Japanese users prefer iso-2022-jp to euc-japan or
454 ;; shift_jis, however iso-8859-1 should be used when
455 ;; there are only ASCII text and Latin-1 characters.
456 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
457 "Preferred coding systems for encoding outgoing messages.
459 More than one suitable coding system may be found for some text.
460 By default, the coding system with the highest priority is used
461 to encode outgoing messages (see `sort-coding-systems'). If this
462 variable is set, it overrides the default priority."
464 :type '(repeat (symbol :tag "Coding system"))
468 (defvar mm-use-find-coding-systems-region
469 (fboundp 'find-coding-systems-region)
470 "Use `find-coding-systems-region' to find proper coding systems.
472 Setting it to nil is useful on Emacsen supporting Unicode if sending
473 mail with multiple parts is preferred to sending a Unicode one.")
475 ;;; Internal variables:
479 (defun mm-mule-charset-to-mime-charset (charset)
480 "Return the MIME charset corresponding to the given Mule CHARSET."
481 (if (and (fboundp 'find-coding-systems-for-charsets)
482 (fboundp 'sort-coding-systems))
483 (let ((css (sort (sort-coding-systems
484 (find-coding-systems-for-charsets (list charset)))
485 'mm-sort-coding-systems-predicate))
487 (while (and (not mime)
489 (when (setq cs (pop css))
490 (setq mime (or (coding-system-get cs :mime-charset)
491 (coding-system-get cs 'mime-charset)))))
493 (let ((alist (mapcar (lambda (cs)
494 (assq cs mm-mime-mule-charset-alist))
495 (sort (mapcar 'car mm-mime-mule-charset-alist)
496 'mm-sort-coding-systems-predicate)))
499 (when (memq charset (cdar alist))
500 (setq out (caar alist)
505 (defun mm-charset-to-coding-system (charset &optional lbt
507 "Return coding-system corresponding to CHARSET.
508 CHARSET is a symbol naming a MIME charset.
509 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
510 used as the line break code type of the coding system.
512 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
513 map undesired charset names to their replacement. This should
514 only be used for decoding, not for encoding."
515 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
516 (when (stringp charset)
517 (setq charset (intern (downcase charset))))
519 (setq charset (intern (format "%s-%s" charset lbt))))
523 ;; Running in a non-MULE environment.
524 ((or (null (mm-get-coding-system-list))
525 (not (fboundp 'coding-system-get)))
527 ;; Check override list quite early. Should only used for decoding, not for
530 (let ((cs (cdr (assq charset mm-charset-override-alist))))
531 (and cs (mm-coding-system-p cs) cs))))
533 ((eq charset 'us-ascii)
535 ;; Check to see whether we can handle this charset. (This depends
536 ;; on there being some coding system matching each `mime-charset'
537 ;; property defined, as there should be.)
538 ((and (mm-coding-system-p charset)
539 ;;; Doing this would potentially weed out incorrect charsets.
541 ;;; (eq charset (coding-system-get charset 'mime-charset))
544 ;; Eval expressions from `mm-charset-eval-alist'
545 ((let* ((el (assq charset mm-charset-eval-alist))
552 (condition-case nil (eval form) (error nil))
553 ;; (message "Failed to eval `%s'" form))
554 (mm-coding-system-p cs)
555 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
557 ;; Translate invalid charsets.
558 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
560 (mm-coding-system-p cs)
562 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
565 ;; Last resort: search the coding system list for entries which
566 ;; have the right mime-charset in case the canonical name isn't
567 ;; defined (though it should be).
569 ;; mm-get-coding-system-list returns a list of cs without lbt.
571 (dolist (c (mm-get-coding-system-list))
573 (eq charset (or (coding-system-get c :mime-charset)
574 (coding-system-get c 'mime-charset))))
577 ;; Warn the user about unknown charset:
578 (if (fboundp 'gnus-message)
579 (gnus-message 7 "Unknown charset: %s" charset)
580 (message "Unknown charset: %s" charset)))
584 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
585 (boundp 'default-enable-multibyte-characters)
586 default-enable-multibyte-characters
587 (fboundp 'set-buffer-multibyte))
588 "True in Emacs with Mule.")
591 (defun mm-enable-multibyte ()
592 "Set the multibyte flag of the current buffer.
593 Only do this if the default value of `enable-multibyte-characters' is
594 non-nil. This is a no-op in XEmacs."
595 (set-buffer-multibyte 'to))
596 (defalias 'mm-enable-multibyte 'ignore))
599 (defun mm-disable-multibyte ()
600 "Unset the multibyte flag of in the current buffer.
601 This is a no-op in XEmacs."
602 (set-buffer-multibyte nil))
603 (defalias 'mm-disable-multibyte 'ignore)))
605 (defun mm-preferred-coding-system (charset)
606 ;; A typo in some Emacs versions.
607 (or (get-charset-property charset 'preferred-coding-system)
608 (get-charset-property charset 'prefered-coding-system)))
610 ;; Mule charsets shouldn't be used.
611 (defsubst mm-guess-charset ()
612 "Guess Mule charset from the language environment."
614 mail-parse-mule-charset ;; cached mule-charset
616 (setq mail-parse-mule-charset
617 (and (boundp 'current-language-environment)
620 (assoc current-language-environment
621 language-info-alist))))))
622 (if (or (not mail-parse-mule-charset)
623 (eq mail-parse-mule-charset 'ascii))
624 (setq mail-parse-mule-charset
625 (or (car (last (assq mail-parse-charset
626 mm-mime-mule-charset-alist)))
629 mail-parse-mule-charset)))
631 (defun mm-charset-after (&optional pos)
632 "Return charset of a character in current buffer at position POS.
633 If POS is nil, it defauls to the current point.
634 If POS is out of range, the value is nil.
635 If the charset is `composition', return the actual one."
636 (let ((char (char-after pos)) charset)
637 (if (< (mm-char-int char) 128)
638 (setq charset 'ascii)
639 ;; charset-after is fake in some Emacsen.
640 (setq charset (and (fboundp 'char-charset) (char-charset char)))
641 (if (eq charset 'composition) ; Mule 4
642 (let ((p (or pos (point))))
643 (cadr (find-charset-region p (1+ p))))
644 (if (and charset (not (memq charset '(ascii eight-bit-control
645 eight-bit-graphic))))
647 (mm-guess-charset))))))
649 (defun mm-mime-charset (charset)
650 "Return the MIME charset corresponding to the given Mule CHARSET."
651 (if (eq charset 'unknown)
652 (error "The message contains non-printable characters, please use attachment"))
653 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
654 ;; This exists in Emacs 20.
656 (and (mm-preferred-coding-system charset)
657 (or (coding-system-get
658 (mm-preferred-coding-system charset) :mime-charset)
660 (mm-preferred-coding-system charset) 'mime-charset)))
661 (and (eq charset 'ascii)
663 (mm-preferred-coding-system charset)
664 (mm-mule-charset-to-mime-charset charset))
665 ;; This is for XEmacs.
666 (mm-mule-charset-to-mime-charset charset)))
668 (if (fboundp 'delete-dups)
669 (defalias 'mm-delete-duplicates 'delete-dups)
670 (defun mm-delete-duplicates (list)
671 "Destructively remove `equal' duplicates from LIST.
672 Store the result in LIST and return it. LIST must be a proper list.
673 Of several `equal' occurrences of an element in LIST, the first
676 This is a compatibility function for Emacsen without `delete-dups'."
677 ;; Code from `subr.el' in Emacs 22:
680 (setcdr tail (delete (car tail) (cdr tail)))
681 (setq tail (cdr tail))))
684 ;; Fixme: This is used in places when it should be testing the
685 ;; default multibyteness. See mm-default-multibyte-p.
687 (if (and (not (featurep 'xemacs))
688 (boundp 'enable-multibyte-characters))
689 (defun mm-multibyte-p ()
690 "Non-nil if multibyte is enabled in the current buffer."
691 enable-multibyte-characters)
692 (defun mm-multibyte-p () (featurep 'mule))))
694 (defun mm-default-multibyte-p ()
695 "Return non-nil if the session is multibyte.
696 This affects whether coding conversion should be attempted generally."
698 (if (boundp 'default-enable-multibyte-characters)
699 default-enable-multibyte-characters
702 (defun mm-iso-8859-x-to-15-region (&optional b e)
703 (if (fboundp 'char-charset)
704 (let (charset item c inconvertible)
706 (if e (narrow-to-region b e))
707 (goto-char (point-min))
708 (skip-chars-forward "\0-\177")
711 ((not (setq item (assq (char-charset (setq c (char-after)))
712 mm-iso-8859-x-to-15-table)))
714 ((memq c (cdr (cdr item)))
715 (setq inconvertible t)
718 (insert-before-markers (prog1 (+ c (car (cdr item)))
720 (skip-chars-forward "\0-\177")))
721 (not inconvertible))))
723 (defun mm-sort-coding-systems-predicate (a b)
726 ;; Note: invalid entries are dropped silently
727 (and (setq cs (mm-coding-system-p cs))
728 (coding-system-base cs)))
729 mm-coding-system-priorities)))
730 (and (setq a (mm-coding-system-p a))
731 (if (setq b (mm-coding-system-p b))
732 (> (length (memq (coding-system-base a) priorities))
733 (length (memq (coding-system-base b) priorities)))
737 (autoload 'latin-unity-massage-name "latin-unity")
738 (autoload 'latin-unity-maybe-remap "latin-unity")
739 (autoload 'latin-unity-representations-feasible-region "latin-unity")
740 (autoload 'latin-unity-representations-present-region "latin-unity")
741 (defvar latin-unity-coding-systems)
742 (defvar latin-unity-ucs-list))
744 (defun mm-xemacs-find-mime-charset-1 (begin end)
745 "Determine which MIME charset to use to send region as message.
746 This uses the XEmacs-specific latin-unity package to better handle the
747 case where identical characters from diverse ISO-8859-? character sets
748 can be encoded using a single one of the corresponding coding systems.
750 It treats `mm-coding-system-priorities' as the list of preferred
751 coding systems; a useful example setting for this list in Western
752 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
753 to the very standard Latin 1 coding system, and only move to coding
754 systems that are less supported as is necessary to encode the
755 characters that exist in the buffer.
757 Latin Unity doesn't know about those non-ASCII Roman characters that
758 are available in various East Asian character sets. As such, its
759 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
760 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
761 But this is very much a corner case, so don't worry about it."
762 (let ((systems mm-coding-system-priorities) csets psets curset)
764 ;; Load the Latin Unity library, if available.
765 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
766 (require 'latin-unity))
768 ;; Now, can we use it?
769 (if (featurep 'latin-unity)
771 (setq csets (latin-unity-representations-feasible-region begin end)
772 psets (latin-unity-representations-present-region begin end))
776 ;; Pass back the first coding system in the preferred list
777 ;; that can encode the whole region.
778 (dolist (curset systems)
779 (setq curset (latin-unity-massage-name 'buffer-default curset))
781 ;; If the coding system is a universal coding system, then
782 ;; it can certainly encode all the characters in the region.
783 (if (memq curset latin-unity-ucs-list)
784 (throw 'done (list curset)))
786 ;; If a coding system isn't universal, and isn't in
787 ;; the list that latin unity knows about, we can't
788 ;; decide whether to use it here. Leave that until later
789 ;; in `mm-find-mime-charset-region' function, whence we
791 (unless (memq curset latin-unity-coding-systems)
794 ;; Right, we know about this coding system, and it may
795 ;; conceivably be able to encode all the characters in
797 (if (latin-unity-maybe-remap begin end curset csets psets t)
798 (throw 'done (list curset))))
800 ;; Can't encode using anything from the
801 ;; `mm-coding-system-priorities' list.
802 ;; Leave `mm-find-mime-charset' to do most of the work.
805 ;; Right, latin unity isn't available; let `mm-find-charset-region'
806 ;; take its default action, which equally applies to GNU Emacs.
809 (defmacro mm-xemacs-find-mime-charset (begin end)
810 (when (featurep 'xemacs)
811 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
813 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
814 "Return the MIME charsets needed to encode the region between B and E.
815 nil means ASCII, a single-element list represents an appropriate MIME
816 charset, and a longer list means no appropriate charset."
818 ;; The return possibilities of this function are a mess...
819 (or (and (mm-multibyte-p)
820 mm-use-find-coding-systems-region
821 ;; Find the mime-charset of the most preferred coding
822 ;; system that has one.
823 (let ((systems (find-coding-systems-region b e)))
824 (when mm-coding-system-priorities
826 (sort systems 'mm-sort-coding-systems-predicate)))
827 (setq systems (delq 'compound-text systems))
828 (unless (equal systems '(undecided))
830 (let* ((head (pop systems))
831 (cs (or (coding-system-get head :mime-charset)
832 (coding-system-get head 'mime-charset))))
833 ;; The mime-charset (`x-ctext') of
834 ;; `compound-text' is not in the IANA list. We
835 ;; shouldn't normally use anything here with a
836 ;; mime-charset having an `x-' prefix.
837 ;; Fixme: Allow this to be overridden, since
838 ;; there is existing use of x-ctext.
839 ;; Also people apparently need the coding system
840 ;; `iso-2022-jp-3' (which Mule-UCS defines with
841 ;; mime-charset, though it's not valid).
843 (not (string-match "^[Xx]-" (symbol-name cs)))
844 ;; UTF-16 of any variety is invalid for
845 ;; text parts and, unfortunately, has
846 ;; mime-charset defined both in Mule-UCS
847 ;; and versions of Emacs. (The name
848 ;; might be `mule-utf-16...' or
850 (not (string-match "utf-16" (symbol-name cs))))
852 charsets (list cs))))))
854 ;; If we're XEmacs, and some coding system is appropriate,
855 ;; mm-xemacs-find-mime-charset will return an appropriate list.
856 ;; Otherwise, we'll get nil, and the next setq will get invoked.
857 (setq charsets (mm-xemacs-find-mime-charset b e))
859 ;; We're not multibyte, or a single coding system won't cover it.
861 (mm-delete-duplicates
862 (mapcar 'mm-mime-charset
864 (mm-find-charset-region b e))))))
865 (if (and (> (length charsets) 1)
866 (memq 'iso-8859-15 charsets)
867 (memq 'iso-8859-15 hack-charsets)
868 (save-excursion (mm-iso-8859-x-to-15-region b e)))
869 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
870 mm-iso-8859-15-compatible))
871 (if (and (memq 'iso-2022-jp-2 charsets)
872 (memq 'iso-2022-jp-2 hack-charsets))
873 (setq charsets (delq 'iso-2022-jp charsets)))
874 ;; Attempt to reduce the number of charsets if utf-8 is available.
875 (if (and (featurep 'xemacs)
876 (> (length charsets) 1)
877 (mm-coding-system-p 'utf-8))
878 (let ((mm-coding-system-priorities
879 (cons 'utf-8 mm-coding-system-priorities)))
881 (mm-delete-duplicates
882 (mapcar 'mm-mime-charset
884 (mm-find-charset-region b e)))))))
887 (defmacro mm-with-unibyte-buffer (&rest forms)
888 "Create a temporary buffer, and evaluate FORMS there like `progn'.
889 Use unibyte mode for this."
890 `(let (default-enable-multibyte-characters)
891 (with-temp-buffer ,@forms)))
892 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
893 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
895 (defmacro mm-with-multibyte-buffer (&rest forms)
896 "Create a temporary buffer, and evaluate FORMS there like `progn'.
897 Use multibyte mode for this."
898 `(let ((default-enable-multibyte-characters t))
899 (with-temp-buffer ,@forms)))
900 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
901 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
903 (defmacro mm-with-unibyte-current-buffer (&rest forms)
904 "Evaluate FORMS with current buffer temporarily made unibyte.
905 Also bind `default-enable-multibyte-characters' to nil.
906 Equivalent to `progn' in XEmacs
908 NOTE: Use this macro with caution in multibyte buffers (it is not
909 worth using this macro in unibyte buffers of course). Use of
910 `(set-buffer-multibyte t)', which is run finally, is generally
911 harmful since it is likely to modify existing data in the buffer.
912 For instance, it converts \"\\300\\255\" into \"\\255\" in
914 (let ((multibyte (make-symbol "multibyte"))
915 (buffer (make-symbol "buffer")))
917 (let ((,multibyte enable-multibyte-characters)
918 (,buffer (current-buffer)))
920 (let (default-enable-multibyte-characters)
921 (set-buffer-multibyte nil)
924 (set-buffer-multibyte ,multibyte)))
925 (let (default-enable-multibyte-characters)
927 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
928 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
930 (defmacro mm-with-unibyte (&rest forms)
931 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
932 `(let (default-enable-multibyte-characters)
934 (put 'mm-with-unibyte 'lisp-indent-function 0)
935 (put 'mm-with-unibyte 'edebug-form-spec '(body))
937 (defmacro mm-with-multibyte (&rest forms)
938 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
939 `(let ((default-enable-multibyte-characters t))
941 (put 'mm-with-multibyte 'lisp-indent-function 0)
942 (put 'mm-with-multibyte 'edebug-form-spec '(body))
944 (defun mm-find-charset-region (b e)
945 "Return a list of Emacs charsets in the region B to E."
947 ((and (mm-multibyte-p)
948 (fboundp 'find-charset-region))
949 ;; Remove composition since the base charsets have been included.
950 ;; Remove eight-bit-*, treat them as ascii.
951 (let ((css (find-charset-region b e)))
952 (mapcar (lambda (cs) (setq css (delq cs css)))
953 '(composition eight-bit-control eight-bit-graphic
957 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
960 (narrow-to-region b e)
961 (goto-char (point-min))
962 (skip-chars-forward "\0-\177")
967 (and (boundp 'current-language-environment)
968 (car (last (assq 'charset
969 (assoc current-language-environment
970 language-info-alist))))))
971 (if (eq charset 'ascii) (setq charset nil))
974 (car (last (assq mail-parse-charset
975 mm-mime-mule-charset-alist)))))
976 (list 'ascii (or charset 'latin-iso8859-1)))))))))
978 (defun mm-auto-mode-alist ()
979 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
980 (let ((alist auto-mode-alist)
983 (when (listp (cdar alist))
984 (push (car alist) out))
988 (defvar mm-inhibit-file-name-handlers
989 '(jka-compr-handler image-file-handler)
990 "A list of handlers doing (un)compression (etc) thingies.")
992 (defun mm-insert-file-contents (filename &optional visit beg end replace
994 "Like `insert-file-contents', but only reads in the file.
995 A buffer may be modified in several ways after reading into the buffer due
996 to advanced Emacs features, such as file-name-handlers, format decoding,
997 `find-file-hooks', etc.
998 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
999 This function ensures that none of these modifications will take place."
1000 (let* ((format-alist nil)
1001 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
1002 (default-major-mode 'fundamental-mode)
1003 (enable-local-variables nil)
1004 (after-insert-file-functions nil)
1005 (enable-local-eval nil)
1006 (inhibit-file-name-operation (if inhibit
1007 'insert-file-contents
1008 inhibit-file-name-operation))
1009 (inhibit-file-name-handlers
1011 (append mm-inhibit-file-name-handlers
1012 inhibit-file-name-handlers)
1013 inhibit-file-name-handlers))
1014 (ffh (if (boundp 'find-file-hook)
1017 (val (symbol-value ffh)))
1020 (insert-file-contents filename visit beg end replace)
1023 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1024 "Append the contents of the region to the end of file FILENAME.
1025 When called from a function, expects three arguments,
1026 START, END and FILENAME. START and END are buffer positions
1027 saying what text to write.
1028 Optional fourth argument specifies the coding system to use when
1030 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1031 (let ((coding-system-for-write
1032 (or codesys mm-text-coding-system-for-write
1033 mm-text-coding-system))
1034 (inhibit-file-name-operation (if inhibit
1036 inhibit-file-name-operation))
1037 (inhibit-file-name-handlers
1039 (append mm-inhibit-file-name-handlers
1040 inhibit-file-name-handlers)
1041 inhibit-file-name-handlers)))
1042 (write-region start end filename t 'no-message)
1043 (message "Appended to %s" filename)))
1045 (defun mm-write-region (start end filename &optional append visit lockname
1046 coding-system inhibit)
1048 "Like `write-region'.
1049 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1050 (let ((coding-system-for-write
1051 (or coding-system mm-text-coding-system-for-write
1052 mm-text-coding-system))
1053 (inhibit-file-name-operation (if inhibit
1055 inhibit-file-name-operation))
1056 (inhibit-file-name-handlers
1058 (append mm-inhibit-file-name-handlers
1059 inhibit-file-name-handlers)
1060 inhibit-file-name-handlers)))
1061 (write-region start end filename append visit lockname)))
1063 ;; It is not a MIME function, but some MIME functions use it.
1064 (if (and (fboundp 'make-temp-file)
1066 (let ((def (symbol-function 'make-temp-file)))
1067 (and (byte-code-function-p def)
1068 (setq def (if (fboundp 'compiled-function-arglist)
1070 (eval (list 'compiled-function-arglist def))
1073 (eq (nth 3 def) 'suffix)))))
1074 (defalias 'mm-make-temp-file 'make-temp-file)
1075 ;; Stolen (and modified for XEmacs) from Emacs 22.
1076 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1077 "Create a temporary file.
1078 The returned file name (created by appending some random characters at the end
1079 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1080 is guaranteed to point to a newly created empty file.
1081 You can then use `write-region' to write new data into the file.
1083 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1085 If SUFFIX is non-nil, add that at the end of the file name."
1086 (let ((umask (default-file-modes))
1090 ;; Create temp files with strict access rights. It's easy to
1091 ;; loosen them later, whereas it's impossible to close the
1092 ;; time-window of loose permissions otherwise.
1093 (set-default-file-modes 448)
1094 (while (condition-case err
1100 (if (fboundp 'temp-directory)
1103 temporary-file-directory))))
1105 (setq file (concat file suffix)))
1107 (make-directory file)
1108 (if (featurep 'xemacs)
1109 ;; NOTE: This is unsafe if XEmacs users
1110 ;; don't use a secure temp directory.
1111 (if (file-exists-p file)
1112 (signal 'file-already-exists
1113 (list "File exists" file))
1114 (write-region "" nil file nil 'silent))
1115 (write-region "" nil file nil 'silent
1118 (file-already-exists t)
1119 ;; The XEmacs version of `make-directory' issues
1121 (file-error (or (and (featurep 'xemacs)
1122 (file-exists-p file))
1123 (signal (car err) (cdr err)))))
1124 ;; the file was somehow created by someone else between
1125 ;; `make-temp-name' and `write-region', let's try again.
1129 (set-default-file-modes umask)))))
1131 (defun mm-image-load-path (&optional package)
1133 (dolist (path load-path (nreverse result))
1136 (setq dir (concat (file-name-directory
1137 (directory-file-name path))
1138 "etc/images/" (or package "gnus/")))))
1140 (push path result))))
1142 ;; Fixme: This doesn't look useful where it's used.
1143 (if (fboundp 'detect-coding-region)
1144 (defun mm-detect-coding-region (start end)
1145 "Like `detect-coding-region' except returning the best one."
1146 (let ((coding-systems
1147 (detect-coding-region start end)))
1148 (or (car-safe coding-systems)
1150 (defun mm-detect-coding-region (start end)
1151 (let ((point (point)))
1153 (skip-chars-forward "\0-\177" end)
1155 (if (eq (point) end) 'ascii (mm-guess-charset))
1156 (goto-char point)))))
1158 (if (fboundp 'coding-system-get)
1159 (defun mm-detect-mime-charset-region (start end)
1160 "Detect MIME charset of the text in the region between START and END."
1161 (let ((cs (mm-detect-coding-region start end)))
1162 (or (coding-system-get cs :mime-charset)
1163 (coding-system-get cs 'mime-charset))))
1164 (defun mm-detect-mime-charset-region (start end)
1165 "Detect MIME charset of the text in the region between START and END."
1166 (let ((cs (mm-detect-coding-region start end)))
1170 (unless (fboundp 'coding-system-to-mime-charset)
1171 (defalias 'coding-system-to-mime-charset 'ignore)))
1173 (defun mm-coding-system-to-mime-charset (coding-system)
1174 "Return the MIME charset corresponding to CODING-SYSTEM.
1175 To make this function work with XEmacs, the APEL package is required."
1177 (or (and (fboundp 'coding-system-get)
1178 (or (coding-system-get coding-system :mime-charset)
1179 (coding-system-get coding-system 'mime-charset)))
1180 (and (featurep 'xemacs)
1181 (or (and (fboundp 'coding-system-to-mime-charset)
1182 (not (eq (symbol-function 'coding-system-to-mime-charset)
1184 (and (condition-case nil
1187 (fboundp 'coding-system-to-mime-charset)))
1188 (coding-system-to-mime-charset coding-system)))))
1191 (require 'jka-compr))
1193 (defun mm-decompress-buffer (filename &optional inplace force)
1194 "Decompress buffer's contents, depending on jka-compr.
1195 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1196 agrees with `jka-compr-compression-info-list', decompression is done.
1197 Signal an error if FORCE is neither nil nor t and compressed data are
1198 not decompressed because `auto-compression-mode' is disabled.
1199 If INPLACE is nil, return decompressed data or nil without modifying
1200 the buffer. Otherwise, replace the buffer's contents with the
1201 decompressed data. The buffer's multibyteness must be turned off."
1204 (prog1 t (require 'jka-compr))
1205 (and (fboundp 'jka-compr-installed-p)
1206 (jka-compr-installed-p))))
1207 (let ((info (jka-compr-get-compression-info filename)))
1209 (unless (or (memq force (list nil t))
1210 (jka-compr-installed-p))
1212 (let ((prog (jka-compr-info-uncompress-program info))
1213 (args (jka-compr-info-uncompress-args info))
1214 (msg (format "%s %s..."
1215 (jka-compr-info-uncompress-message info)
1217 (err-file (jka-compr-make-temp-name))
1218 (cur (current-buffer))
1219 (coding-system-for-read mm-binary-coding-system)
1220 (coding-system-for-write mm-binary-coding-system)
1223 (mm-with-unibyte-buffer
1224 (insert-buffer-substring cur)
1227 (unless (memq (apply 'call-process-region
1228 (point-min) (point-max)
1229 prog t (list t err-file) nil args)
1230 jka-compr-acceptable-retval-list)
1234 (delete "" (split-string
1236 (insert-file-contents err-file)
1242 (format "Error while executing \"%s %s < %s\""
1243 prog (mapconcat 'identity args " ")
1245 (setq retval (buffer-string)))
1247 (setq err-msg (error-message-string err)))))
1248 (when (file-exists-p err-file)
1249 (ignore-errors (jka-compr-delete-temp-file err-file)))
1252 (delete-region (point-min) (point-max))
1255 (message "%s" (or err-msg (concat msg "done")))
1259 (unless (fboundp 'coding-system-name)
1260 (defalias 'coding-system-name 'ignore))
1261 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1262 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1263 (unless (fboundp 'find-operation-coding-system)
1264 (defalias 'find-operation-coding-system 'ignore)))
1266 (defun mm-find-buffer-file-coding-system (&optional filename)
1267 "Find coding system used to decode the contents of the current buffer.
1268 This function looks for the coding system magic cookie or examines the
1269 coding system specified by `file-coding-system-alist' being associated
1270 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1271 gzip, bzip2, etc. are allowed."
1273 (setq filename buffer-file-name))
1275 (let ((decomp (unless ;; No worth to examine charset of tar files.
1278 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1280 (mm-decompress-buffer filename nil t))))
1282 (set-buffer (let (default-enable-multibyte-characters)
1283 (generate-new-buffer " *temp*")))
1285 (setq filename (file-name-sans-extension filename)))
1286 (goto-char (point-min))
1289 ((boundp 'set-auto-coding-function) ;; Emacs
1291 (or (funcall (symbol-value 'set-auto-coding-function)
1292 filename (- (point-max) (point-min)))
1293 (car (find-operation-coding-system 'insert-file-contents
1295 (let (auto-coding-alist)
1297 (funcall (symbol-value 'set-auto-coding-function)
1298 nil (- (point-max) (point-min)))
1300 ((featurep 'file-coding) ;; XEmacs
1301 (let ((case-fold-search t)
1302 (end (point-at-eol))
1305 (and (re-search-forward "-\\*-+[\t ]*" end t)
1307 (setq start (match-end 0))
1308 (re-search-forward "[\t ]*-+\\*-" end t))
1310 (setq end (match-beginning 0))
1312 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1314 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1316 (find-coding-system (setq codesys
1317 (intern (match-string 1))))
1319 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1322 (setq start (match-end 0))
1323 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1325 (setq end (match-beginning 0))
1328 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1330 (find-coding-system (setq codesys
1331 (intern (match-string 1))))
1334 (goto-char (point-min))
1335 (setq case-fold-search nil)
1336 (re-search-forward "^;;;coding system: "
1337 ;;(+ (point-min) 3000) t))
1339 (looking-at "[^\t\n\r ]+")
1341 (setq codesys (intern (match-string 0))))
1345 (find-file-coding-system-for-read-from-filename
1347 (coding-system-name (coding-system-base codesys)))))))
1349 (kill-buffer (current-buffer)))))))
1353 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1354 ;;; mm-util.el ends here