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)
103 (multibyte-string-p . ignore)
104 (insert-byte . insert-char)
105 (multibyte-char-to-unibyte . identity)
107 . (lambda (buffer-name)
108 "Returns non-nil if a buffer named BUFFER-NAME gets a special frame."
109 (and special-display-function
110 (or (and (member buffer-name special-display-buffer-names) t)
111 (cdr (assoc buffer-name special-display-buffer-names))
113 (dolist (elem special-display-regexps)
115 (string-match elem buffer-name)
119 (string-match (car elem) buffer-name)
120 (throw 'return (cdr elem))))))))))))
122 (defalias 'mm-string-to-multibyte
126 ((fboundp 'string-to-multibyte)
127 'string-to-multibyte)
130 "Return a multibyte string with the same individual chars as string."
132 (lambda (ch) (mm-string-as-multibyte (char-to-string ch)))
136 (defalias 'mm-char-or-char-int-p
138 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
139 ((fboundp 'char-valid-p) 'char-valid-p)
142 ;; Fixme: This seems always to be used to read a MIME charset, so it
143 ;; should be re-named and fixed (in Emacs) to offer completion only on
144 ;; proper charset names (base coding systems which have a
145 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
147 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
148 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
149 ;; Actually, there should be an `mm-coding-system-mime-charset'.
151 (defalias 'mm-read-coding-system
153 ((fboundp 'read-coding-system)
154 (if (and (featurep 'xemacs)
155 (<= (string-to-number emacs-version) 21.1))
156 (lambda (prompt &optional default-coding-system)
157 (read-coding-system prompt))
158 'read-coding-system))
159 (t (lambda (prompt &optional default-coding-system)
160 "Prompt the user for a coding system."
162 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
163 mm-mime-mule-charset-alist)))))))
165 (defvar mm-coding-system-list nil)
166 (defun mm-get-coding-system-list ()
167 "Get the coding system list."
168 (or mm-coding-system-list
169 (setq mm-coding-system-list (mm-coding-system-list))))
171 (defun mm-coding-system-p (cs)
172 "Return non-nil if CS is a symbol naming a coding system.
173 In XEmacs, also return non-nil if CS is a coding system object.
174 If CS is available, return CS itself in Emacs, and return a coding
175 system object in XEmacs."
176 (if (fboundp 'find-coding-system)
177 (and cs (find-coding-system cs))
178 (if (fboundp 'coding-system-p)
179 (when (coding-system-p cs)
182 (car (memq cs (mm-get-coding-system-list))))))
184 (defun mm-codepage-setup (number &optional alias)
185 "Create a coding system cpNUMBER.
186 The coding system is created using `codepage-setup'. If ALIAS is
187 non-nil, an alias is created and added to
188 `mm-charset-synonym-alist'. If ALIAS is a string, it's used as
189 the alias. Else windows-NUMBER is used."
191 (let ((completion-ignore-case t)
192 (candidates (cp-supported-codepages)))
193 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
194 nil t nil nil "437"))))
196 (setq alias (if (stringp alias)
198 (intern (format "windows-%s" number)))))
199 (let* ((cp (intern (format "cp%s" number))))
200 (unless (mm-coding-system-p cp)
201 (codepage-setup number))
203 ;; Don't add alias if setup of cp failed.
204 (mm-coding-system-p cp))
205 (add-to-list 'mm-charset-synonym-alist (cons alias cp)))))
207 (defvar mm-charset-synonym-alist
209 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
210 ,@(unless (mm-coding-system-p 'x-ctext)
211 '((x-ctext . ctext)))
212 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_ in 8
214 ,@(unless (mm-coding-system-p 'iso-8859-15)
215 '((iso-8859-15 . iso-8859-1)))
216 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
217 ,@(unless (mm-coding-system-p 'big5-hkscs)
218 '((big5-hkscs . big5)))
219 ;; A Microsoft misunderstanding.
220 ,@(when (and (not (mm-coding-system-p 'unicode))
221 (mm-coding-system-p 'utf-16-le))
222 '((unicode . utf-16-le)))
223 ;; A Microsoft misunderstanding.
224 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
225 (if (mm-coding-system-p 'cp949)
226 '((ks_c_5601-1987 . cp949))
227 '((ks_c_5601-1987 . euc-kr))))
228 ;; Windows-31J is Windows Codepage 932.
229 ,@(when (and (not (mm-coding-system-p 'windows-31j))
230 (mm-coding-system-p 'cp932))
231 '((windows-31j . cp932)))
233 "A mapping from unknown or invalid charset names to the real charset names.
235 See `mm-codepage-iso-8859-list' and `mm-codepage-ibm-list'.")
237 (defcustom mm-codepage-iso-8859-list
238 (list 1250 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
239 ;; Outlook users in Czech republic. Use this to allow reading of
240 ;; their e-mails. cp1250 should be defined by M-x codepage-setup
242 '(1252 . 1) ;; Windows-1252 is a superset of iso-8859-1 (West
243 ;; Europe). See also `gnus-article-dumbquotes-map'.
244 '(1254 . 9) ;; Windows-1254 is a superset of iso-8859-9 (Turkish).
245 '(1255 . 8));; Windows-1255 is a superset of iso-8859-8 (Hebrew).
246 "A list of Windows codepage numbers and iso-8859 charset numbers.
248 If an element is a number corresponding to a supported windows
249 codepage, appropriate entries to `mm-charset-synonym-alist' are
250 added by `mm-setup-codepage-iso-8859'. An element may also be a
251 cons cell where the car is a codepage number and the cdr is the
252 corresponding number of an iso-8859 charset."
253 :type '(list (set :inline t
254 (const 1250 :tag "Central and East European")
255 (const (1252 . 1) :tag "West European")
256 (const (1254 . 9) :tag "Turkish")
257 (const (1255 . 8) :tag "Hebrew"))
261 (integer :tag "Windows codepage number")
262 (cons (integer :tag "Windows codepage number")
263 (integer :tag "iso-8859 charset number")))))
264 :version "22.1" ;; Gnus 5.10.9
267 (defcustom mm-codepage-ibm-list
268 (list 437 ;; (US etc.)
272 863 ;; (Canadian French)
277 866 ;; (Cyrillic - Russian)
282 ;; In Emacs 23 (unicode), cp... and ibm... are aliases.
283 ;; Cf. http://thread.gmane.org/v9lkng5nwy.fsf@marauder.physik.uni-ulm.de
284 "List of IBM codepage numbers.
286 The codepage mappings slighly differ between IBM and other vendors.
287 See \"ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/IBM/README.TXT\".
289 If an element is a number corresponding to a supported windows
290 codepage, appropriate entries to `mm-charset-synonym-alist' are
291 added by `mm-setup-codepage-ibm'."
292 :type '(list (set :inline t
293 (const 437 :tag "US etc.")
294 (const 860 :tag "Portugal")
295 (const 861 :tag "Iceland")
296 (const 862 :tag "Israel")
297 (const 863 :tag "Canadian French")
298 (const 865 :tag "Nordic")
300 (const 850 :tag "Latin 1")
301 (const 855 :tag "Cyrillic")
302 (const 866 :tag "Cyrillic - Russian")
303 (const 857 :tag "Turkish")
304 (const 864 :tag "Arabic")
305 (const 869 :tag "Greek")
306 (const 874 :tag "Thai"))
309 (integer :tag "Codepage number")))
310 :version "22.1" ;; Gnus 5.10.9
313 (defun mm-setup-codepage-iso-8859 (&optional list)
314 "Add appropriate entries to `mm-charset-synonym-alist'.
315 Unless LIST is given, `mm-codepage-iso-8859-list' is used."
317 (setq list mm-codepage-iso-8859-list))
319 (let (cp windows iso)
321 (setq cp (intern (format "cp%d" (car i)))
322 windows (intern (format "windows-%d" (car i)))
323 iso (intern (format "iso-8859-%d" (cdr i))))
324 (setq cp (intern (format "cp%d" i))
325 windows (intern (format "windows-%d" i))))
326 (unless (mm-coding-system-p windows)
327 (if (mm-coding-system-p cp)
328 (add-to-list 'mm-charset-synonym-alist (cons windows cp))
329 (add-to-list 'mm-charset-synonym-alist (cons windows iso)))))))
331 (defun mm-setup-codepage-ibm (&optional list)
332 "Add appropriate entries to `mm-charset-synonym-alist'.
333 Unless LIST is given, `mm-codepage-ibm-list' is used."
335 (setq list mm-codepage-ibm-list))
336 (dolist (number list)
337 (let ((ibm (intern (format "ibm%d" number)))
338 (cp (intern (format "cp%d" number))))
339 (when (and (not (mm-coding-system-p ibm))
340 (mm-coding-system-p cp))
341 (add-to-list 'mm-charset-synonym-alist (cons ibm cp))))))
344 (mm-setup-codepage-iso-8859)
345 (mm-setup-codepage-ibm)
347 (defcustom mm-charset-override-alist
348 '((iso-8859-1 . windows-1252)
349 (iso-8859-8 . windows-1255)
350 (iso-8859-9 . windows-1254))
351 "A mapping from undesired charset names to their replacement.
353 You may add pairs like (iso-8859-1 . windows-1252) here,
354 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
355 superset of iso-8859-1."
356 :type '(list (set :inline t
357 (const (iso-8859-1 . windows-1252))
358 (const (iso-8859-8 . windows-1255))
359 (const (iso-8859-9 . windows-1254))
360 (const (undecided . windows-1252)))
363 (cons (symbol :tag "From charset")
364 (symbol :tag "To charset"))))
365 :version "22.1" ;; Gnus 5.10.9
368 (defcustom mm-charset-eval-alist
369 (if (featurep 'xemacs)
370 nil ;; I don't know what would be useful for XEmacs.
371 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
372 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
373 (windows-1250 . (mm-codepage-setup 1250 t))
374 (windows-1251 . (mm-codepage-setup 1251 t))
375 (windows-1253 . (mm-codepage-setup 1253 t))
376 (windows-1257 . (mm-codepage-setup 1257 t))))
377 "An alist of (CHARSET . FORM) pairs.
378 If an article is encoded in an unknown CHARSET, FORM is
379 evaluated. This allows to load additional libraries providing
380 charsets on demand. If supported by your Emacs version, you
381 could use `autoload-coding-system' here."
382 :version "22.1" ;; Gnus 5.10.9
383 :type '(list (set :inline t
384 (const (windows-1250 . (mm-codepage-setup 1250 t)))
385 (const (windows-1251 . (mm-codepage-setup 1251 t)))
386 (const (windows-1253 . (mm-codepage-setup 1253 t)))
387 (const (windows-1257 . (mm-codepage-setup 1257 t)))
388 (const (cp850 . (mm-codepage-setup 850 nil))))
391 (cons (symbol :tag "charset")
392 (symbol :tag "form"))))
395 (defvar mm-binary-coding-system
397 ((mm-coding-system-p 'binary) 'binary)
398 ((mm-coding-system-p 'no-conversion) 'no-conversion)
400 "100% binary coding system.")
402 (defvar mm-text-coding-system
403 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
404 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
405 (and (mm-coding-system-p 'raw-text) 'raw-text))
406 mm-binary-coding-system)
407 "Text-safe coding system (For removing ^M).")
409 (defvar mm-text-coding-system-for-write nil
410 "Text coding system for write.")
412 (defvar mm-auto-save-coding-system
414 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
415 (if (memq system-type '(windows-nt ms-dos ms-windows))
416 (if (mm-coding-system-p 'utf-8-emacs-dos)
417 'utf-8-emacs-dos mm-binary-coding-system)
419 ((mm-coding-system-p 'emacs-mule)
420 (if (memq system-type '(windows-nt ms-dos ms-windows))
421 (if (mm-coding-system-p 'emacs-mule-dos)
422 'emacs-mule-dos mm-binary-coding-system)
424 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
425 (t mm-binary-coding-system))
426 "Coding system of auto save file.")
428 (defvar mm-universal-coding-system mm-auto-save-coding-system
429 "The universal coding system.")
431 ;; Fixme: some of the cars here aren't valid MIME charsets. That
432 ;; should only matter with XEmacs, though.
433 (defvar mm-mime-mule-charset-alist
435 (iso-8859-1 latin-iso8859-1)
436 (iso-8859-2 latin-iso8859-2)
437 (iso-8859-3 latin-iso8859-3)
438 (iso-8859-4 latin-iso8859-4)
439 (iso-8859-5 cyrillic-iso8859-5)
440 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
441 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
442 ;; charset is koi8-r, not iso-8859-5.
443 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
444 (iso-8859-6 arabic-iso8859-6)
445 (iso-8859-7 greek-iso8859-7)
446 (iso-8859-8 hebrew-iso8859-8)
447 (iso-8859-9 latin-iso8859-9)
448 (iso-8859-14 latin-iso8859-14)
449 (iso-8859-15 latin-iso8859-15)
450 (viscii vietnamese-viscii-lower)
451 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
452 (euc-kr korean-ksc5601)
453 (gb2312 chinese-gb2312)
454 (big5 chinese-big5-1 chinese-big5-2)
456 (thai-tis620 thai-tis620)
457 (windows-1251 cyrillic-iso8859-5)
458 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
459 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
460 latin-jisx0201 japanese-jisx0208-1978
461 chinese-gb2312 japanese-jisx0208
462 korean-ksc5601 japanese-jisx0212)
463 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
464 latin-jisx0201 japanese-jisx0208-1978
465 chinese-gb2312 japanese-jisx0208
466 korean-ksc5601 japanese-jisx0212
467 chinese-cns11643-1 chinese-cns11643-2)
468 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
469 cyrillic-iso8859-5 greek-iso8859-7
470 latin-jisx0201 japanese-jisx0208-1978
471 chinese-gb2312 japanese-jisx0208
472 korean-ksc5601 japanese-jisx0212
473 chinese-cns11643-1 chinese-cns11643-2
474 chinese-cns11643-3 chinese-cns11643-4
475 chinese-cns11643-5 chinese-cns11643-6
477 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
478 japanese-jisx0213-1 japanese-jisx0213-2)
479 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
480 ,(cond ((fboundp 'unicode-precedence-list)
481 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
482 (unicode-precedence-list)))))
483 ((or (not (fboundp 'charsetp)) ;; non-Mule case
484 (charsetp 'unicode-a)
485 (not (mm-coding-system-p 'mule-utf-8)))
486 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
487 (t ;; If we have utf-8 we're in Mule 5+.
490 (coding-system-get 'mule-utf-8 'safe-charsets))))))
491 "Alist of MIME-charset/MULE-charsets.")
493 (defun mm-enrich-utf-8-by-mule-ucs ()
494 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
495 This function will run when the `un-define' module is loaded under
496 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
497 with Mule charsets. It is completely useless for Emacs."
498 (when (boundp 'unicode-basic-translation-charset-order-list)
504 'unicode-basic-translation-charset-order-list))))
505 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
508 (setq mm-mime-mule-charset-alist
509 (nconc mm-mime-mule-charset-alist
510 (list (cons 'utf-8 val))))))
513 ;; Correct by construction, but should be unnecessary for Emacs:
514 (if (featurep 'xemacs)
515 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
516 (when (and (fboundp 'coding-system-list)
517 (fboundp 'sort-coding-systems))
518 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
522 mime (or (coding-system-get cs :mime-charset) ; Emacs 23 (unicode)
523 (coding-system-get cs 'mime-charset)))
525 (not (eq t (setq mule
526 (coding-system-get cs 'safe-charsets))))
527 (not (assq mime alist)))
528 (push (cons mime (delq 'ascii mule)) alist)))
529 (setq mm-mime-mule-charset-alist (nreverse alist)))))
531 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
532 "A list of special charsets.
533 Valid elements include:
534 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
535 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
538 (defvar mm-iso-8859-15-compatible
539 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
540 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
541 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
543 (defvar mm-iso-8859-x-to-15-table
544 (and (fboundp 'coding-system-p)
545 (mm-coding-system-p 'iso-8859-15)
548 (if (mm-coding-system-p (car cs))
549 (let ((c (string-to-char
550 (decode-coding-string "\341" (car cs)))))
551 (cons (char-charset c)
554 (decode-coding-string "\341" 'iso-8859-15)) c)
555 (string-to-list (decode-coding-string (car (cdr cs))
558 mm-iso-8859-15-compatible))
559 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
561 (defcustom mm-coding-system-priorities
562 (if (boundp 'current-language-environment)
563 (let ((lang (symbol-value 'current-language-environment)))
564 (cond ((string= lang "Japanese")
565 ;; Japanese users prefer iso-2022-jp to euc-japan or
566 ;; shift_jis, however iso-8859-1 should be used when
567 ;; there are only ASCII text and Latin-1 characters.
568 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
569 "Preferred coding systems for encoding outgoing messages.
571 More than one suitable coding system may be found for some text.
572 By default, the coding system with the highest priority is used
573 to encode outgoing messages (see `sort-coding-systems'). If this
574 variable is set, it overrides the default priority."
576 :type '(repeat (symbol :tag "Coding system"))
580 (defvar mm-use-find-coding-systems-region
581 (fboundp 'find-coding-systems-region)
582 "Use `find-coding-systems-region' to find proper coding systems.
584 Setting it to nil is useful on Emacsen supporting Unicode if sending
585 mail with multiple parts is preferred to sending a Unicode one.")
587 ;;; Internal variables:
591 (defun mm-mule-charset-to-mime-charset (charset)
592 "Return the MIME charset corresponding to the given Mule CHARSET."
593 (if (and (fboundp 'find-coding-systems-for-charsets)
594 (fboundp 'sort-coding-systems))
595 (let ((css (sort (sort-coding-systems
596 (find-coding-systems-for-charsets (list charset)))
597 'mm-sort-coding-systems-predicate))
599 (while (and (not mime)
601 (when (setq cs (pop css))
602 (setq mime (or (coding-system-get cs :mime-charset)
603 (coding-system-get cs 'mime-charset)))))
605 (let ((alist (mapcar (lambda (cs)
606 (assq cs mm-mime-mule-charset-alist))
607 (sort (mapcar 'car mm-mime-mule-charset-alist)
608 'mm-sort-coding-systems-predicate)))
611 (when (memq charset (cdar alist))
612 (setq out (caar alist)
617 (defun mm-charset-to-coding-system (charset &optional lbt
619 "Return coding-system corresponding to CHARSET.
620 CHARSET is a symbol naming a MIME charset.
621 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
622 used as the line break code type of the coding system.
624 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
625 map undesired charset names to their replacement. This should
626 only be used for decoding, not for encoding."
627 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
628 (when (stringp charset)
629 (setq charset (intern (downcase charset))))
631 (setq charset (intern (format "%s-%s" charset lbt))))
635 ;; Running in a non-MULE environment.
636 ((or (null (mm-get-coding-system-list))
637 (not (fboundp 'coding-system-get)))
639 ;; Check override list quite early. Should only used for decoding, not for
642 (let ((cs (cdr (assq charset mm-charset-override-alist))))
643 (and cs (mm-coding-system-p cs) cs))))
645 ((eq charset 'us-ascii)
647 ;; Check to see whether we can handle this charset. (This depends
648 ;; on there being some coding system matching each `mime-charset'
649 ;; property defined, as there should be.)
650 ((and (mm-coding-system-p charset)
651 ;;; Doing this would potentially weed out incorrect charsets.
653 ;;; (eq charset (coding-system-get charset 'mime-charset))
656 ;; Eval expressions from `mm-charset-eval-alist'
657 ((let* ((el (assq charset mm-charset-eval-alist))
664 (condition-case nil (eval form) (error nil))
665 ;; (message "Failed to eval `%s'" form))
666 (mm-coding-system-p cs)
667 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
669 ;; Translate invalid charsets.
670 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
672 (mm-coding-system-p cs)
674 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
677 ;; Last resort: search the coding system list for entries which
678 ;; have the right mime-charset in case the canonical name isn't
679 ;; defined (though it should be).
681 ;; mm-get-coding-system-list returns a list of cs without lbt.
683 (dolist (c (mm-get-coding-system-list))
685 (eq charset (or (coding-system-get c :mime-charset)
686 (coding-system-get c 'mime-charset))))
689 ;; Warn the user about unknown charset:
690 (if (fboundp 'gnus-message)
691 (gnus-message 7 "Unknown charset: %s" charset)
692 (message "Unknown charset: %s" charset)))
696 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
697 (boundp 'default-enable-multibyte-characters)
698 default-enable-multibyte-characters
699 (fboundp 'set-buffer-multibyte))
700 "True in Emacs with Mule.")
703 (defun mm-enable-multibyte ()
704 "Set the multibyte flag of the current buffer.
705 Only do this if the default value of `enable-multibyte-characters' is
706 non-nil. This is a no-op in XEmacs."
707 (set-buffer-multibyte 'to))
708 (defalias 'mm-enable-multibyte 'ignore))
711 (defun mm-disable-multibyte ()
712 "Unset the multibyte flag of in the current buffer.
713 This is a no-op in XEmacs."
714 (set-buffer-multibyte nil))
715 (defalias 'mm-disable-multibyte 'ignore)))
717 (defun mm-preferred-coding-system (charset)
718 ;; A typo in some Emacs versions.
719 (or (get-charset-property charset 'preferred-coding-system)
720 (get-charset-property charset 'prefered-coding-system)))
722 ;; Mule charsets shouldn't be used.
723 (defsubst mm-guess-charset ()
724 "Guess Mule charset from the language environment."
726 mail-parse-mule-charset ;; cached mule-charset
728 (setq mail-parse-mule-charset
729 (and (boundp 'current-language-environment)
732 (assoc current-language-environment
733 language-info-alist))))))
734 (if (or (not mail-parse-mule-charset)
735 (eq mail-parse-mule-charset 'ascii))
736 (setq mail-parse-mule-charset
737 (or (car (last (assq mail-parse-charset
738 mm-mime-mule-charset-alist)))
741 mail-parse-mule-charset)))
743 (defun mm-charset-after (&optional pos)
744 "Return charset of a character in current buffer at position POS.
745 If POS is nil, it defauls to the current point.
746 If POS is out of range, the value is nil.
747 If the charset is `composition', return the actual one."
748 (let ((char (char-after pos)) charset)
749 (if (< (mm-char-int char) 128)
750 (setq charset 'ascii)
751 ;; charset-after is fake in some Emacsen.
752 (setq charset (and (fboundp 'char-charset) (char-charset char)))
753 (if (eq charset 'composition) ; Mule 4
754 (let ((p (or pos (point))))
755 (cadr (find-charset-region p (1+ p))))
756 (if (and charset (not (memq charset '(ascii eight-bit-control
757 eight-bit-graphic))))
759 (mm-guess-charset))))))
761 (defun mm-mime-charset (charset)
762 "Return the MIME charset corresponding to the given Mule CHARSET."
763 (if (eq charset 'unknown)
764 (error "The message contains non-printable characters, please use attachment"))
765 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
766 ;; This exists in Emacs 20.
768 (and (mm-preferred-coding-system charset)
769 (or (coding-system-get
770 (mm-preferred-coding-system charset) :mime-charset)
772 (mm-preferred-coding-system charset) 'mime-charset)))
773 (and (eq charset 'ascii)
775 (mm-preferred-coding-system charset)
776 (mm-mule-charset-to-mime-charset charset))
777 ;; This is for XEmacs.
778 (mm-mule-charset-to-mime-charset charset)))
780 (if (fboundp 'delete-dups)
781 (defalias 'mm-delete-duplicates 'delete-dups)
782 (defun mm-delete-duplicates (list)
783 "Destructively remove `equal' duplicates from LIST.
784 Store the result in LIST and return it. LIST must be a proper list.
785 Of several `equal' occurrences of an element in LIST, the first
788 This is a compatibility function for Emacsen without `delete-dups'."
789 ;; Code from `subr.el' in Emacs 22:
792 (setcdr tail (delete (car tail) (cdr tail)))
793 (setq tail (cdr tail))))
796 ;; Fixme: This is used in places when it should be testing the
797 ;; default multibyteness. See mm-default-multibyte-p.
799 (if (and (not (featurep 'xemacs))
800 (boundp 'enable-multibyte-characters))
801 (defun mm-multibyte-p ()
802 "Non-nil if multibyte is enabled in the current buffer."
803 enable-multibyte-characters)
804 (defun mm-multibyte-p () (featurep 'mule))))
806 (defun mm-default-multibyte-p ()
807 "Return non-nil if the session is multibyte.
808 This affects whether coding conversion should be attempted generally."
810 (if (boundp 'default-enable-multibyte-characters)
811 default-enable-multibyte-characters
814 (defun mm-iso-8859-x-to-15-region (&optional b e)
815 (if (fboundp 'char-charset)
816 (let (charset item c inconvertible)
818 (if e (narrow-to-region b e))
819 (goto-char (point-min))
820 (skip-chars-forward "\0-\177")
823 ((not (setq item (assq (char-charset (setq c (char-after)))
824 mm-iso-8859-x-to-15-table)))
826 ((memq c (cdr (cdr item)))
827 (setq inconvertible t)
830 (insert-before-markers (prog1 (+ c (car (cdr item)))
832 (skip-chars-forward "\0-\177")))
833 (not inconvertible))))
835 (defun mm-sort-coding-systems-predicate (a b)
838 ;; Note: invalid entries are dropped silently
839 (and (setq cs (mm-coding-system-p cs))
840 (coding-system-base cs)))
841 mm-coding-system-priorities)))
842 (and (setq a (mm-coding-system-p a))
843 (if (setq b (mm-coding-system-p b))
844 (> (length (memq (coding-system-base a) priorities))
845 (length (memq (coding-system-base b) priorities)))
849 (autoload 'latin-unity-massage-name "latin-unity")
850 (autoload 'latin-unity-maybe-remap "latin-unity")
851 (autoload 'latin-unity-representations-feasible-region "latin-unity")
852 (autoload 'latin-unity-representations-present-region "latin-unity")
853 (defvar latin-unity-coding-systems)
854 (defvar latin-unity-ucs-list))
856 (defun mm-xemacs-find-mime-charset-1 (begin end)
857 "Determine which MIME charset to use to send region as message.
858 This uses the XEmacs-specific latin-unity package to better handle the
859 case where identical characters from diverse ISO-8859-? character sets
860 can be encoded using a single one of the corresponding coding systems.
862 It treats `mm-coding-system-priorities' as the list of preferred
863 coding systems; a useful example setting for this list in Western
864 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
865 to the very standard Latin 1 coding system, and only move to coding
866 systems that are less supported as is necessary to encode the
867 characters that exist in the buffer.
869 Latin Unity doesn't know about those non-ASCII Roman characters that
870 are available in various East Asian character sets. As such, its
871 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
872 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
873 But this is very much a corner case, so don't worry about it."
874 (let ((systems mm-coding-system-priorities) csets psets curset)
876 ;; Load the Latin Unity library, if available.
877 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
878 (require 'latin-unity))
880 ;; Now, can we use it?
881 (if (featurep 'latin-unity)
883 (setq csets (latin-unity-representations-feasible-region begin end)
884 psets (latin-unity-representations-present-region begin end))
888 ;; Pass back the first coding system in the preferred list
889 ;; that can encode the whole region.
890 (dolist (curset systems)
891 (setq curset (latin-unity-massage-name 'buffer-default curset))
893 ;; If the coding system is a universal coding system, then
894 ;; it can certainly encode all the characters in the region.
895 (if (memq curset latin-unity-ucs-list)
896 (throw 'done (list curset)))
898 ;; If a coding system isn't universal, and isn't in
899 ;; the list that latin unity knows about, we can't
900 ;; decide whether to use it here. Leave that until later
901 ;; in `mm-find-mime-charset-region' function, whence we
903 (unless (memq curset latin-unity-coding-systems)
906 ;; Right, we know about this coding system, and it may
907 ;; conceivably be able to encode all the characters in
909 (if (latin-unity-maybe-remap begin end curset csets psets t)
910 (throw 'done (list curset))))
912 ;; Can't encode using anything from the
913 ;; `mm-coding-system-priorities' list.
914 ;; Leave `mm-find-mime-charset' to do most of the work.
917 ;; Right, latin unity isn't available; let `mm-find-charset-region'
918 ;; take its default action, which equally applies to GNU Emacs.
921 (defmacro mm-xemacs-find-mime-charset (begin end)
922 (when (featurep 'xemacs)
923 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
925 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
926 "Return the MIME charsets needed to encode the region between B and E.
927 nil means ASCII, a single-element list represents an appropriate MIME
928 charset, and a longer list means no appropriate charset."
930 ;; The return possibilities of this function are a mess...
931 (or (and (mm-multibyte-p)
932 mm-use-find-coding-systems-region
933 ;; Find the mime-charset of the most preferred coding
934 ;; system that has one.
935 (let ((systems (find-coding-systems-region b e)))
936 (when mm-coding-system-priorities
938 (sort systems 'mm-sort-coding-systems-predicate)))
939 (setq systems (delq 'compound-text systems))
940 (unless (equal systems '(undecided))
942 (let* ((head (pop systems))
943 (cs (or (coding-system-get head :mime-charset)
944 (coding-system-get head 'mime-charset))))
945 ;; The mime-charset (`x-ctext') of
946 ;; `compound-text' is not in the IANA list. We
947 ;; shouldn't normally use anything here with a
948 ;; mime-charset having an `x-' prefix.
949 ;; Fixme: Allow this to be overridden, since
950 ;; there is existing use of x-ctext.
951 ;; Also people apparently need the coding system
952 ;; `iso-2022-jp-3' (which Mule-UCS defines with
953 ;; mime-charset, though it's not valid).
955 (not (string-match "^[Xx]-" (symbol-name cs)))
956 ;; UTF-16 of any variety is invalid for
957 ;; text parts and, unfortunately, has
958 ;; mime-charset defined both in Mule-UCS
959 ;; and versions of Emacs. (The name
960 ;; might be `mule-utf-16...' or
962 (not (string-match "utf-16" (symbol-name cs))))
964 charsets (list cs))))))
966 ;; If we're XEmacs, and some coding system is appropriate,
967 ;; mm-xemacs-find-mime-charset will return an appropriate list.
968 ;; Otherwise, we'll get nil, and the next setq will get invoked.
969 (setq charsets (mm-xemacs-find-mime-charset b e))
971 ;; We're not multibyte, or a single coding system won't cover it.
973 (mm-delete-duplicates
974 (mapcar 'mm-mime-charset
976 (mm-find-charset-region b e))))))
977 (if (and (> (length charsets) 1)
978 (memq 'iso-8859-15 charsets)
979 (memq 'iso-8859-15 hack-charsets)
980 (save-excursion (mm-iso-8859-x-to-15-region b e)))
981 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
982 mm-iso-8859-15-compatible))
983 (if (and (memq 'iso-2022-jp-2 charsets)
984 (memq 'iso-2022-jp-2 hack-charsets))
985 (setq charsets (delq 'iso-2022-jp charsets)))
986 ;; Attempt to reduce the number of charsets if utf-8 is available.
987 (if (and (featurep 'xemacs)
988 (> (length charsets) 1)
989 (mm-coding-system-p 'utf-8))
990 (let ((mm-coding-system-priorities
991 (cons 'utf-8 mm-coding-system-priorities)))
993 (mm-delete-duplicates
994 (mapcar 'mm-mime-charset
996 (mm-find-charset-region b e)))))))
999 (defmacro mm-with-unibyte-buffer (&rest forms)
1000 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1001 Use unibyte mode for this."
1002 `(let (default-enable-multibyte-characters)
1003 (with-temp-buffer ,@forms)))
1004 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
1005 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
1007 (defmacro mm-with-multibyte-buffer (&rest forms)
1008 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1009 Use multibyte mode for this."
1010 `(let ((default-enable-multibyte-characters t))
1011 (with-temp-buffer ,@forms)))
1012 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
1013 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
1015 (defmacro mm-with-unibyte-current-buffer (&rest forms)
1016 "Evaluate FORMS with current buffer temporarily made unibyte.
1017 Also bind `default-enable-multibyte-characters' to nil.
1018 Equivalent to `progn' in XEmacs
1020 NOTE: Use this macro with caution in multibyte buffers (it is not
1021 worth using this macro in unibyte buffers of course). Use of
1022 `(set-buffer-multibyte t)', which is run finally, is generally
1023 harmful since it is likely to modify existing data in the buffer.
1024 For instance, it converts \"\\300\\255\" into \"\\255\" in
1025 Emacs 23 (unicode)."
1026 (let ((multibyte (make-symbol "multibyte"))
1027 (buffer (make-symbol "buffer")))
1029 (let ((,multibyte enable-multibyte-characters)
1030 (,buffer (current-buffer)))
1032 (let (default-enable-multibyte-characters)
1033 (set-buffer-multibyte nil)
1035 (set-buffer ,buffer)
1036 (set-buffer-multibyte ,multibyte)))
1037 (let (default-enable-multibyte-characters)
1039 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
1040 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
1042 (defmacro mm-with-unibyte (&rest forms)
1043 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
1044 `(let (default-enable-multibyte-characters)
1046 (put 'mm-with-unibyte 'lisp-indent-function 0)
1047 (put 'mm-with-unibyte 'edebug-form-spec '(body))
1049 (defmacro mm-with-multibyte (&rest forms)
1050 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
1051 `(let ((default-enable-multibyte-characters t))
1053 (put 'mm-with-multibyte 'lisp-indent-function 0)
1054 (put 'mm-with-multibyte 'edebug-form-spec '(body))
1056 (defun mm-find-charset-region (b e)
1057 "Return a list of Emacs charsets in the region B to E."
1059 ((and (mm-multibyte-p)
1060 (fboundp 'find-charset-region))
1061 ;; Remove composition since the base charsets have been included.
1062 ;; Remove eight-bit-*, treat them as ascii.
1063 (let ((css (find-charset-region b e)))
1064 (mapcar (lambda (cs) (setq css (delq cs css)))
1065 '(composition eight-bit-control eight-bit-graphic
1069 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
1072 (narrow-to-region b e)
1073 (goto-char (point-min))
1074 (skip-chars-forward "\0-\177")
1079 (and (boundp 'current-language-environment)
1080 (car (last (assq 'charset
1081 (assoc current-language-environment
1082 language-info-alist))))))
1083 (if (eq charset 'ascii) (setq charset nil))
1086 (car (last (assq mail-parse-charset
1087 mm-mime-mule-charset-alist)))))
1088 (list 'ascii (or charset 'latin-iso8859-1)))))))))
1090 (defun mm-auto-mode-alist ()
1091 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
1092 (let ((alist auto-mode-alist)
1095 (when (listp (cdar alist))
1096 (push (car alist) out))
1100 (defvar mm-inhibit-file-name-handlers
1101 '(jka-compr-handler image-file-handler)
1102 "A list of handlers doing (un)compression (etc) thingies.")
1104 (defun mm-insert-file-contents (filename &optional visit beg end replace
1106 "Like `insert-file-contents', but only reads in the file.
1107 A buffer may be modified in several ways after reading into the buffer due
1108 to advanced Emacs features, such as file-name-handlers, format decoding,
1109 `find-file-hooks', etc.
1110 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
1111 This function ensures that none of these modifications will take place."
1112 (let* ((format-alist nil)
1113 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
1114 (default-major-mode 'fundamental-mode)
1115 (enable-local-variables nil)
1116 (after-insert-file-functions nil)
1117 (enable-local-eval nil)
1118 (inhibit-file-name-operation (if inhibit
1119 'insert-file-contents
1120 inhibit-file-name-operation))
1121 (inhibit-file-name-handlers
1123 (append mm-inhibit-file-name-handlers
1124 inhibit-file-name-handlers)
1125 inhibit-file-name-handlers))
1126 (ffh (if (boundp 'find-file-hook)
1129 (val (symbol-value ffh)))
1132 (insert-file-contents filename visit beg end replace)
1135 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1136 "Append the contents of the region to the end of file FILENAME.
1137 When called from a function, expects three arguments,
1138 START, END and FILENAME. START and END are buffer positions
1139 saying what text to write.
1140 Optional fourth argument specifies the coding system to use when
1142 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1143 (let ((coding-system-for-write
1144 (or codesys mm-text-coding-system-for-write
1145 mm-text-coding-system))
1146 (inhibit-file-name-operation (if inhibit
1148 inhibit-file-name-operation))
1149 (inhibit-file-name-handlers
1151 (append mm-inhibit-file-name-handlers
1152 inhibit-file-name-handlers)
1153 inhibit-file-name-handlers)))
1154 (write-region start end filename t 'no-message)
1155 (message "Appended to %s" filename)))
1157 (defun mm-write-region (start end filename &optional append visit lockname
1158 coding-system inhibit)
1160 "Like `write-region'.
1161 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1162 (let ((coding-system-for-write
1163 (or coding-system mm-text-coding-system-for-write
1164 mm-text-coding-system))
1165 (inhibit-file-name-operation (if inhibit
1167 inhibit-file-name-operation))
1168 (inhibit-file-name-handlers
1170 (append mm-inhibit-file-name-handlers
1171 inhibit-file-name-handlers)
1172 inhibit-file-name-handlers)))
1173 (write-region start end filename append visit lockname)))
1175 ;; It is not a MIME function, but some MIME functions use it.
1176 (if (and (fboundp 'make-temp-file)
1178 (let ((def (symbol-function 'make-temp-file)))
1179 (and (byte-code-function-p def)
1180 (setq def (if (fboundp 'compiled-function-arglist)
1182 (eval (list 'compiled-function-arglist def))
1185 (eq (nth 3 def) 'suffix)))))
1186 (defalias 'mm-make-temp-file 'make-temp-file)
1187 ;; Stolen (and modified for XEmacs) from Emacs 22.
1188 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1189 "Create a temporary file.
1190 The returned file name (created by appending some random characters at the end
1191 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1192 is guaranteed to point to a newly created empty file.
1193 You can then use `write-region' to write new data into the file.
1195 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1197 If SUFFIX is non-nil, add that at the end of the file name."
1198 (let ((umask (default-file-modes))
1202 ;; Create temp files with strict access rights. It's easy to
1203 ;; loosen them later, whereas it's impossible to close the
1204 ;; time-window of loose permissions otherwise.
1205 (set-default-file-modes 448)
1206 (while (condition-case err
1212 (if (fboundp 'temp-directory)
1215 temporary-file-directory))))
1217 (setq file (concat file suffix)))
1219 (make-directory file)
1220 ;; NOTE: This is unsafe if Emacs 20
1221 ;; users and XEmacs users don't use
1222 ;; a secure temp directory.
1223 (gmm-write-region "" nil file nil 'silent
1226 (file-already-exists t)
1227 ;; The XEmacs version of `make-directory' issues
1229 (file-error (or (and (featurep 'xemacs)
1230 (file-exists-p file))
1231 (signal (car err) (cdr err)))))
1232 ;; the file was somehow created by someone else between
1233 ;; `make-temp-name' and `write-region', let's try again.
1237 (set-default-file-modes umask)))))
1239 (defun mm-image-load-path (&optional package)
1241 (dolist (path load-path (nreverse result))
1244 (setq dir (concat (file-name-directory
1245 (directory-file-name path))
1246 "etc/images/" (or package "gnus/")))))
1248 (push path result))))
1250 ;; Fixme: This doesn't look useful where it's used.
1251 (if (fboundp 'detect-coding-region)
1252 (defun mm-detect-coding-region (start end)
1253 "Like `detect-coding-region' except returning the best one."
1254 (let ((coding-systems
1255 (detect-coding-region start end)))
1256 (or (car-safe coding-systems)
1258 (defun mm-detect-coding-region (start end)
1259 (let ((point (point)))
1261 (skip-chars-forward "\0-\177" end)
1263 (if (eq (point) end) 'ascii (mm-guess-charset))
1264 (goto-char point)))))
1266 (if (fboundp 'coding-system-get)
1267 (defun mm-detect-mime-charset-region (start end)
1268 "Detect MIME charset of the text in the region between START and END."
1269 (let ((cs (mm-detect-coding-region start end)))
1270 (or (coding-system-get cs :mime-charset)
1271 (coding-system-get cs 'mime-charset))))
1272 (defun mm-detect-mime-charset-region (start end)
1273 "Detect MIME charset of the text in the region between START and END."
1274 (let ((cs (mm-detect-coding-region start end)))
1278 (unless (fboundp 'coding-system-to-mime-charset)
1279 (defalias 'coding-system-to-mime-charset 'ignore)))
1281 (defun mm-coding-system-to-mime-charset (coding-system)
1282 "Return the MIME charset corresponding to CODING-SYSTEM.
1283 To make this function work with XEmacs, the APEL package is required."
1285 (or (and (fboundp 'coding-system-get)
1286 (or (coding-system-get coding-system :mime-charset)
1287 (coding-system-get coding-system 'mime-charset)))
1288 (and (featurep 'xemacs)
1289 (or (and (fboundp 'coding-system-to-mime-charset)
1290 (not (eq (symbol-function 'coding-system-to-mime-charset)
1292 (and (condition-case nil
1295 (fboundp 'coding-system-to-mime-charset)))
1296 (coding-system-to-mime-charset coding-system)))))
1299 (require 'jka-compr))
1301 (defun mm-decompress-buffer (filename &optional inplace force)
1302 "Decompress buffer's contents, depending on jka-compr.
1303 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1304 agrees with `jka-compr-compression-info-list', decompression is done.
1305 Signal an error if FORCE is neither nil nor t and compressed data are
1306 not decompressed because `auto-compression-mode' is disabled.
1307 If INPLACE is nil, return decompressed data or nil without modifying
1308 the buffer. Otherwise, replace the buffer's contents with the
1309 decompressed data. The buffer's multibyteness must be turned off."
1312 (prog1 t (require 'jka-compr))
1313 (and (fboundp 'jka-compr-installed-p)
1314 (jka-compr-installed-p))))
1315 (let ((info (jka-compr-get-compression-info filename)))
1317 (unless (or (memq force (list nil t))
1318 (jka-compr-installed-p))
1320 (let ((prog (jka-compr-info-uncompress-program info))
1321 (args (jka-compr-info-uncompress-args info))
1322 (msg (format "%s %s..."
1323 (jka-compr-info-uncompress-message info)
1325 (err-file (jka-compr-make-temp-name))
1326 (cur (current-buffer))
1327 (coding-system-for-read mm-binary-coding-system)
1328 (coding-system-for-write mm-binary-coding-system)
1331 (mm-with-unibyte-buffer
1332 (insert-buffer-substring cur)
1335 (unless (memq (apply 'call-process-region
1336 (point-min) (point-max)
1337 prog t (list t err-file) nil args)
1338 jka-compr-acceptable-retval-list)
1342 (delete "" (split-string
1344 (insert-file-contents err-file)
1350 (format "Error while executing \"%s %s < %s\""
1351 prog (mapconcat 'identity args " ")
1353 (setq retval (buffer-string)))
1355 (setq err-msg (error-message-string err)))))
1356 (when (file-exists-p err-file)
1357 (ignore-errors (jka-compr-delete-temp-file err-file)))
1360 (delete-region (point-min) (point-max))
1363 (message "%s" (or err-msg (concat msg "done")))
1367 (unless (fboundp 'coding-system-name)
1368 (defalias 'coding-system-name 'ignore))
1369 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1370 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1371 (unless (fboundp 'find-operation-coding-system)
1372 (defalias 'find-operation-coding-system 'ignore)))
1374 (defun mm-find-buffer-file-coding-system (&optional filename)
1375 "Find coding system used to decode the contents of the current buffer.
1376 This function looks for the coding system magic cookie or examines the
1377 coding system specified by `file-coding-system-alist' being associated
1378 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1379 gzip, bzip2, etc. are allowed."
1381 (setq filename buffer-file-name))
1383 (let ((decomp (unless ;; No worth to examine charset of tar files.
1386 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1388 (mm-decompress-buffer filename nil t))))
1390 (set-buffer (let (default-enable-multibyte-characters)
1391 (generate-new-buffer " *temp*")))
1393 (setq filename (file-name-sans-extension filename)))
1394 (goto-char (point-min))
1397 ((boundp 'set-auto-coding-function) ;; Emacs
1399 (or (funcall (symbol-value 'set-auto-coding-function)
1400 filename (- (point-max) (point-min)))
1401 (car (find-operation-coding-system 'insert-file-contents
1403 (let (auto-coding-alist)
1405 (funcall (symbol-value 'set-auto-coding-function)
1406 nil (- (point-max) (point-min)))
1408 ((featurep 'file-coding) ;; XEmacs
1409 (let ((case-fold-search t)
1410 (end (point-at-eol))
1413 (and (re-search-forward "-\\*-+[\t ]*" end t)
1415 (setq start (match-end 0))
1416 (re-search-forward "[\t ]*-+\\*-" end t))
1418 (setq end (match-beginning 0))
1420 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1422 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1424 (find-coding-system (setq codesys
1425 (intern (match-string 1))))
1427 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1430 (setq start (match-end 0))
1431 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1433 (setq end (match-beginning 0))
1436 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1438 (find-coding-system (setq codesys
1439 (intern (match-string 1))))
1442 (goto-char (point-min))
1443 (setq case-fold-search nil)
1444 (re-search-forward "^;;;coding system: "
1445 ;;(+ (point-min) 3000) t))
1447 (looking-at "[^\t\n\r ]+")
1449 (setq codesys (intern (match-string 0))))
1453 (find-file-coding-system-for-read-from-filename
1455 (coding-system-name (coding-system-base codesys)))))))
1457 (kill-buffer (current-buffer)))))))
1461 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1462 ;;; mm-util.el ends here