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, 2007 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 (defcustom 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)))
232 ;; ISO8859-1 is ISO-8859-1
233 ,@(when (and (not (mm-coding-system-p 'iso8859-1))
234 (mm-coding-system-p 'iso-8859-1))
235 '((iso8859-1 . iso-8859-1)))
237 "A mapping from unknown or invalid charset names to the real charset names.
239 See `mm-codepage-iso-8859-list' and `mm-codepage-ibm-list'."
241 :convert-widget (lambda (widget)
242 (apply 'widget-convert (widget-type widget)
243 (eval (car (widget-get widget :args)))))
245 (symbol :tag "Synonym")
246 ,(if (get 'coding-system 'widget-type)
248 '(symbol :tag "Coding system")))))
251 (defcustom mm-codepage-iso-8859-list
252 (list 1250 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
253 ;; Outlook users in Czech republic. Use this to allow reading of
254 ;; their e-mails. cp1250 should be defined by M-x codepage-setup
256 '(1252 . 1) ;; Windows-1252 is a superset of iso-8859-1 (West
257 ;; Europe). See also `gnus-article-dumbquotes-map'.
258 '(1254 . 9) ;; Windows-1254 is a superset of iso-8859-9 (Turkish).
259 '(1255 . 8));; Windows-1255 is a superset of iso-8859-8 (Hebrew).
260 "A list of Windows codepage numbers and iso-8859 charset numbers.
262 If an element is a number corresponding to a supported windows
263 codepage, appropriate entries to `mm-charset-synonym-alist' are
264 added by `mm-setup-codepage-iso-8859'. An element may also be a
265 cons cell where the car is a codepage number and the cdr is the
266 corresponding number of an iso-8859 charset."
267 :type '(list (set :inline t
268 (const 1250 :tag "Central and East European")
269 (const (1252 . 1) :tag "West European")
270 (const (1254 . 9) :tag "Turkish")
271 (const (1255 . 8) :tag "Hebrew"))
275 (integer :tag "Windows codepage number")
276 (cons (integer :tag "Windows codepage number")
277 (integer :tag "iso-8859 charset number")))))
278 :version "22.1" ;; Gnus 5.10.9
281 (defcustom mm-codepage-ibm-list
282 (list 437 ;; (US etc.)
286 863 ;; (Canadian French)
291 866 ;; (Cyrillic - Russian)
296 ;; In Emacs 23 (unicode), cp... and ibm... are aliases.
297 ;; Cf. http://thread.gmane.org/v9lkng5nwy.fsf@marauder.physik.uni-ulm.de
298 "List of IBM codepage numbers.
300 The codepage mappings slighly differ between IBM and other vendors.
301 See \"ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/IBM/README.TXT\".
303 If an element is a number corresponding to a supported windows
304 codepage, appropriate entries to `mm-charset-synonym-alist' are
305 added by `mm-setup-codepage-ibm'."
306 :type '(list (set :inline t
307 (const 437 :tag "US etc.")
308 (const 860 :tag "Portugal")
309 (const 861 :tag "Iceland")
310 (const 862 :tag "Israel")
311 (const 863 :tag "Canadian French")
312 (const 865 :tag "Nordic")
314 (const 850 :tag "Latin 1")
315 (const 855 :tag "Cyrillic")
316 (const 866 :tag "Cyrillic - Russian")
317 (const 857 :tag "Turkish")
318 (const 864 :tag "Arabic")
319 (const 869 :tag "Greek")
320 (const 874 :tag "Thai"))
323 (integer :tag "Codepage number")))
324 :version "22.1" ;; Gnus 5.10.9
327 (defun mm-setup-codepage-iso-8859 (&optional list)
328 "Add appropriate entries to `mm-charset-synonym-alist'.
329 Unless LIST is given, `mm-codepage-iso-8859-list' is used."
331 (setq list mm-codepage-iso-8859-list))
333 (let (cp windows iso)
335 (setq cp (intern (format "cp%d" (car i)))
336 windows (intern (format "windows-%d" (car i)))
337 iso (intern (format "iso-8859-%d" (cdr i))))
338 (setq cp (intern (format "cp%d" i))
339 windows (intern (format "windows-%d" i))))
340 (unless (mm-coding-system-p windows)
341 (if (mm-coding-system-p cp)
342 (add-to-list 'mm-charset-synonym-alist (cons windows cp))
343 (add-to-list 'mm-charset-synonym-alist (cons windows iso)))))))
345 (defun mm-setup-codepage-ibm (&optional list)
346 "Add appropriate entries to `mm-charset-synonym-alist'.
347 Unless LIST is given, `mm-codepage-ibm-list' is used."
349 (setq list mm-codepage-ibm-list))
350 (dolist (number list)
351 (let ((ibm (intern (format "ibm%d" number)))
352 (cp (intern (format "cp%d" number))))
353 (when (and (not (mm-coding-system-p ibm))
354 (mm-coding-system-p cp))
355 (add-to-list 'mm-charset-synonym-alist (cons ibm cp))))))
358 (mm-setup-codepage-iso-8859)
359 (mm-setup-codepage-ibm)
361 (defcustom mm-charset-override-alist
362 '((iso-8859-1 . windows-1252)
363 (iso-8859-8 . windows-1255)
364 (iso-8859-9 . windows-1254))
365 "A mapping from undesired charset names to their replacement.
367 You may add pairs like (iso-8859-1 . windows-1252) here,
368 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
369 superset of iso-8859-1."
370 :type '(list (set :inline t
371 (const (iso-8859-1 . windows-1252))
372 (const (iso-8859-8 . windows-1255))
373 (const (iso-8859-9 . windows-1254))
374 (const (undecided . windows-1252)))
377 (cons (symbol :tag "From charset")
378 (symbol :tag "To charset"))))
379 :version "22.1" ;; Gnus 5.10.9
382 (defcustom mm-charset-eval-alist
383 (if (featurep 'xemacs)
384 nil ;; I don't know what would be useful for XEmacs.
385 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
386 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
387 (windows-1250 . (mm-codepage-setup 1250 t))
388 (windows-1251 . (mm-codepage-setup 1251 t))
389 (windows-1253 . (mm-codepage-setup 1253 t))
390 (windows-1257 . (mm-codepage-setup 1257 t))))
391 "An alist of (CHARSET . FORM) pairs.
392 If an article is encoded in an unknown CHARSET, FORM is
393 evaluated. This allows to load additional libraries providing
394 charsets on demand. If supported by your Emacs version, you
395 could use `autoload-coding-system' here."
396 :version "22.1" ;; Gnus 5.10.9
397 :type '(list (set :inline t
398 (const (windows-1250 . (mm-codepage-setup 1250 t)))
399 (const (windows-1251 . (mm-codepage-setup 1251 t)))
400 (const (windows-1253 . (mm-codepage-setup 1253 t)))
401 (const (windows-1257 . (mm-codepage-setup 1257 t)))
402 (const (cp850 . (mm-codepage-setup 850 nil))))
405 (cons (symbol :tag "charset")
406 (symbol :tag "form"))))
409 (defvar mm-binary-coding-system
411 ((mm-coding-system-p 'binary) 'binary)
412 ((mm-coding-system-p 'no-conversion) 'no-conversion)
414 "100% binary coding system.")
416 (defvar mm-text-coding-system
417 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
418 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
419 (and (mm-coding-system-p 'raw-text) 'raw-text))
420 mm-binary-coding-system)
421 "Text-safe coding system (For removing ^M).")
423 (defvar mm-text-coding-system-for-write nil
424 "Text coding system for write.")
426 (defvar mm-auto-save-coding-system
428 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
429 (if (memq system-type '(windows-nt ms-dos ms-windows))
430 (if (mm-coding-system-p 'utf-8-emacs-dos)
431 'utf-8-emacs-dos mm-binary-coding-system)
433 ((mm-coding-system-p 'emacs-mule)
434 (if (memq system-type '(windows-nt ms-dos ms-windows))
435 (if (mm-coding-system-p 'emacs-mule-dos)
436 'emacs-mule-dos mm-binary-coding-system)
438 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
439 (t mm-binary-coding-system))
440 "Coding system of auto save file.")
442 (defvar mm-universal-coding-system mm-auto-save-coding-system
443 "The universal coding system.")
445 ;; Fixme: some of the cars here aren't valid MIME charsets. That
446 ;; should only matter with XEmacs, though.
447 (defvar mm-mime-mule-charset-alist
449 (iso-8859-1 latin-iso8859-1)
450 (iso-8859-2 latin-iso8859-2)
451 (iso-8859-3 latin-iso8859-3)
452 (iso-8859-4 latin-iso8859-4)
453 (iso-8859-5 cyrillic-iso8859-5)
454 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
455 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
456 ;; charset is koi8-r, not iso-8859-5.
457 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
458 (iso-8859-6 arabic-iso8859-6)
459 (iso-8859-7 greek-iso8859-7)
460 (iso-8859-8 hebrew-iso8859-8)
461 (iso-8859-9 latin-iso8859-9)
462 (iso-8859-14 latin-iso8859-14)
463 (iso-8859-15 latin-iso8859-15)
464 (viscii vietnamese-viscii-lower)
465 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
466 (euc-kr korean-ksc5601)
467 (gb2312 chinese-gb2312)
468 (big5 chinese-big5-1 chinese-big5-2)
470 (thai-tis620 thai-tis620)
471 (windows-1251 cyrillic-iso8859-5)
472 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
473 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
474 latin-jisx0201 japanese-jisx0208-1978
475 chinese-gb2312 japanese-jisx0208
476 korean-ksc5601 japanese-jisx0212)
477 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
478 latin-jisx0201 japanese-jisx0208-1978
479 chinese-gb2312 japanese-jisx0208
480 korean-ksc5601 japanese-jisx0212
481 chinese-cns11643-1 chinese-cns11643-2)
482 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
483 cyrillic-iso8859-5 greek-iso8859-7
484 latin-jisx0201 japanese-jisx0208-1978
485 chinese-gb2312 japanese-jisx0208
486 korean-ksc5601 japanese-jisx0212
487 chinese-cns11643-1 chinese-cns11643-2
488 chinese-cns11643-3 chinese-cns11643-4
489 chinese-cns11643-5 chinese-cns11643-6
491 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
492 japanese-jisx0213-1 japanese-jisx0213-2)
493 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
494 ,(cond ((fboundp 'unicode-precedence-list)
495 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
496 (unicode-precedence-list)))))
497 ((or (not (fboundp 'charsetp)) ;; non-Mule case
498 (charsetp 'unicode-a)
499 (not (mm-coding-system-p 'mule-utf-8)))
500 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
501 (t ;; If we have utf-8 we're in Mule 5+.
504 (coding-system-get 'mule-utf-8 'safe-charsets))))))
505 "Alist of MIME-charset/MULE-charsets.")
507 (defun mm-enrich-utf-8-by-mule-ucs ()
508 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
509 This function will run when the `un-define' module is loaded under
510 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
511 with Mule charsets. It is completely useless for Emacs."
512 (when (boundp 'unicode-basic-translation-charset-order-list)
518 'unicode-basic-translation-charset-order-list))))
519 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
522 (setq mm-mime-mule-charset-alist
523 (nconc mm-mime-mule-charset-alist
524 (list (cons 'utf-8 val))))))
527 ;; Correct by construction, but should be unnecessary for Emacs:
528 (if (featurep 'xemacs)
529 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
530 (when (and (fboundp 'coding-system-list)
531 (fboundp 'sort-coding-systems))
532 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
536 mime (or (coding-system-get cs :mime-charset) ; Emacs 23 (unicode)
537 (coding-system-get cs 'mime-charset)))
539 (not (eq t (setq mule
540 (coding-system-get cs 'safe-charsets))))
541 (not (assq mime alist)))
542 (push (cons mime (delq 'ascii mule)) alist)))
543 (setq mm-mime-mule-charset-alist (nreverse alist)))))
545 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
546 "A list of special charsets.
547 Valid elements include:
548 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
549 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
552 (defvar mm-iso-8859-15-compatible
553 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
554 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
555 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
557 (defvar mm-iso-8859-x-to-15-table
558 (and (fboundp 'coding-system-p)
559 (mm-coding-system-p 'iso-8859-15)
562 (if (mm-coding-system-p (car cs))
563 (let ((c (string-to-char
564 (decode-coding-string "\341" (car cs)))))
565 (cons (char-charset c)
568 (decode-coding-string "\341" 'iso-8859-15)) c)
569 (string-to-list (decode-coding-string (car (cdr cs))
572 mm-iso-8859-15-compatible))
573 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
575 (defcustom mm-coding-system-priorities
576 (if (boundp 'current-language-environment)
577 (let ((lang (symbol-value 'current-language-environment)))
578 (cond ((string= lang "Japanese")
579 ;; Japanese users prefer iso-2022-jp to euc-japan or
580 ;; shift_jis, however iso-8859-1 should be used when
581 ;; there are only ASCII text and Latin-1 characters.
582 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
583 "Preferred coding systems for encoding outgoing messages.
585 More than one suitable coding system may be found for some text.
586 By default, the coding system with the highest priority is used
587 to encode outgoing messages (see `sort-coding-systems'). If this
588 variable is set, it overrides the default priority."
590 :type '(repeat (symbol :tag "Coding system"))
594 (defvar mm-use-find-coding-systems-region
595 (fboundp 'find-coding-systems-region)
596 "Use `find-coding-systems-region' to find proper coding systems.
598 Setting it to nil is useful on Emacsen supporting Unicode if sending
599 mail with multiple parts is preferred to sending a Unicode one.")
601 ;;; Internal variables:
605 (defun mm-mule-charset-to-mime-charset (charset)
606 "Return the MIME charset corresponding to the given Mule CHARSET."
607 (if (and (fboundp 'find-coding-systems-for-charsets)
608 (fboundp 'sort-coding-systems))
609 (let ((css (sort (sort-coding-systems
610 (find-coding-systems-for-charsets (list charset)))
611 'mm-sort-coding-systems-predicate))
613 (while (and (not mime)
615 (when (setq cs (pop css))
616 (setq mime (or (coding-system-get cs :mime-charset)
617 (coding-system-get cs 'mime-charset)))))
619 (let ((alist (mapcar (lambda (cs)
620 (assq cs mm-mime-mule-charset-alist))
621 (sort (mapcar 'car mm-mime-mule-charset-alist)
622 'mm-sort-coding-systems-predicate)))
625 (when (memq charset (cdar alist))
626 (setq out (caar alist)
631 (defun mm-charset-to-coding-system (charset &optional lbt
633 "Return coding-system corresponding to CHARSET.
634 CHARSET is a symbol naming a MIME charset.
635 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
636 used as the line break code type of the coding system.
638 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
639 map undesired charset names to their replacement. This should
640 only be used for decoding, not for encoding."
641 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
642 (when (stringp charset)
643 (setq charset (intern (downcase charset))))
645 (setq charset (intern (format "%s-%s" charset lbt))))
649 ;; Running in a non-MULE environment.
650 ((or (null (mm-get-coding-system-list))
651 (not (fboundp 'coding-system-get)))
653 ;; Check override list quite early. Should only used for decoding, not for
656 (let ((cs (cdr (assq charset mm-charset-override-alist))))
657 (and cs (mm-coding-system-p cs) cs))))
659 ((eq charset 'us-ascii)
661 ;; Check to see whether we can handle this charset. (This depends
662 ;; on there being some coding system matching each `mime-charset'
663 ;; property defined, as there should be.)
664 ((and (mm-coding-system-p charset)
665 ;;; Doing this would potentially weed out incorrect charsets.
667 ;;; (eq charset (coding-system-get charset 'mime-charset))
670 ;; Eval expressions from `mm-charset-eval-alist'
671 ((let* ((el (assq charset mm-charset-eval-alist))
678 (condition-case nil (eval form) (error nil))
679 ;; (message "Failed to eval `%s'" form))
680 (mm-coding-system-p cs)
681 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
683 ;; Translate invalid charsets.
684 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
686 (mm-coding-system-p cs)
688 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
691 ;; Last resort: search the coding system list for entries which
692 ;; have the right mime-charset in case the canonical name isn't
693 ;; defined (though it should be).
695 ;; mm-get-coding-system-list returns a list of cs without lbt.
697 (dolist (c (mm-get-coding-system-list))
699 (eq charset (or (coding-system-get c :mime-charset)
700 (coding-system-get c 'mime-charset))))
703 ;; Warn the user about unknown charset:
704 (if (fboundp 'gnus-message)
705 (gnus-message 7 "Unknown charset: %s" charset)
706 (message "Unknown charset: %s" charset)))
710 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
711 (boundp 'default-enable-multibyte-characters)
712 default-enable-multibyte-characters
713 (fboundp 'set-buffer-multibyte))
714 "True in Emacs with Mule.")
717 (defun mm-enable-multibyte ()
718 "Set the multibyte flag of the current buffer.
719 Only do this if the default value of `enable-multibyte-characters' is
720 non-nil. This is a no-op in XEmacs."
721 (set-buffer-multibyte 'to))
722 (defalias 'mm-enable-multibyte 'ignore))
725 (defun mm-disable-multibyte ()
726 "Unset the multibyte flag of in the current buffer.
727 This is a no-op in XEmacs."
728 (set-buffer-multibyte nil))
729 (defalias 'mm-disable-multibyte 'ignore)))
731 (defun mm-preferred-coding-system (charset)
732 ;; A typo in some Emacs versions.
733 (or (get-charset-property charset 'preferred-coding-system)
734 (get-charset-property charset 'prefered-coding-system)))
736 ;; Mule charsets shouldn't be used.
737 (defsubst mm-guess-charset ()
738 "Guess Mule charset from the language environment."
740 mail-parse-mule-charset ;; cached mule-charset
742 (setq mail-parse-mule-charset
743 (and (boundp 'current-language-environment)
746 (assoc current-language-environment
747 language-info-alist))))))
748 (if (or (not mail-parse-mule-charset)
749 (eq mail-parse-mule-charset 'ascii))
750 (setq mail-parse-mule-charset
751 (or (car (last (assq mail-parse-charset
752 mm-mime-mule-charset-alist)))
755 mail-parse-mule-charset)))
757 (defun mm-charset-after (&optional pos)
758 "Return charset of a character in current buffer at position POS.
759 If POS is nil, it defauls to the current point.
760 If POS is out of range, the value is nil.
761 If the charset is `composition', return the actual one."
762 (let ((char (char-after pos)) charset)
763 (if (< (mm-char-int char) 128)
764 (setq charset 'ascii)
765 ;; charset-after is fake in some Emacsen.
766 (setq charset (and (fboundp 'char-charset) (char-charset char)))
767 (if (eq charset 'composition) ; Mule 4
768 (let ((p (or pos (point))))
769 (cadr (find-charset-region p (1+ p))))
770 (if (and charset (not (memq charset '(ascii eight-bit-control
771 eight-bit-graphic))))
773 (mm-guess-charset))))))
775 (defun mm-mime-charset (charset)
776 "Return the MIME charset corresponding to the given Mule CHARSET."
777 (if (eq charset 'unknown)
778 (error "The message contains non-printable characters, please use attachment"))
779 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
780 ;; This exists in Emacs 20.
782 (and (mm-preferred-coding-system charset)
783 (or (coding-system-get
784 (mm-preferred-coding-system charset) :mime-charset)
786 (mm-preferred-coding-system charset) 'mime-charset)))
787 (and (eq charset 'ascii)
789 (mm-preferred-coding-system charset)
790 (mm-mule-charset-to-mime-charset charset))
791 ;; This is for XEmacs.
792 (mm-mule-charset-to-mime-charset charset)))
794 (if (fboundp 'delete-dups)
795 (defalias 'mm-delete-duplicates 'delete-dups)
796 (defun mm-delete-duplicates (list)
797 "Destructively remove `equal' duplicates from LIST.
798 Store the result in LIST and return it. LIST must be a proper list.
799 Of several `equal' occurrences of an element in LIST, the first
802 This is a compatibility function for Emacsen without `delete-dups'."
803 ;; Code from `subr.el' in Emacs 22:
806 (setcdr tail (delete (car tail) (cdr tail)))
807 (setq tail (cdr tail))))
810 ;; Fixme: This is used in places when it should be testing the
811 ;; default multibyteness. See mm-default-multibyte-p.
813 (if (and (not (featurep 'xemacs))
814 (boundp 'enable-multibyte-characters))
815 (defun mm-multibyte-p ()
816 "Non-nil if multibyte is enabled in the current buffer."
817 enable-multibyte-characters)
818 (defun mm-multibyte-p () (featurep 'mule))))
820 (defun mm-default-multibyte-p ()
821 "Return non-nil if the session is multibyte.
822 This affects whether coding conversion should be attempted generally."
824 (if (boundp 'default-enable-multibyte-characters)
825 default-enable-multibyte-characters
828 (defun mm-iso-8859-x-to-15-region (&optional b e)
829 (if (fboundp 'char-charset)
830 (let (charset item c inconvertible)
832 (if e (narrow-to-region b e))
833 (goto-char (point-min))
834 (skip-chars-forward "\0-\177")
837 ((not (setq item (assq (char-charset (setq c (char-after)))
838 mm-iso-8859-x-to-15-table)))
840 ((memq c (cdr (cdr item)))
841 (setq inconvertible t)
844 (insert-before-markers (prog1 (+ c (car (cdr item)))
846 (skip-chars-forward "\0-\177")))
847 (not inconvertible))))
849 (defun mm-sort-coding-systems-predicate (a b)
852 ;; Note: invalid entries are dropped silently
853 (and (setq cs (mm-coding-system-p cs))
854 (coding-system-base cs)))
855 mm-coding-system-priorities)))
856 (and (setq a (mm-coding-system-p a))
857 (if (setq b (mm-coding-system-p b))
858 (> (length (memq (coding-system-base a) priorities))
859 (length (memq (coding-system-base b) priorities)))
863 (autoload 'latin-unity-massage-name "latin-unity")
864 (autoload 'latin-unity-maybe-remap "latin-unity")
865 (autoload 'latin-unity-representations-feasible-region "latin-unity")
866 (autoload 'latin-unity-representations-present-region "latin-unity")
867 (defvar latin-unity-coding-systems)
868 (defvar latin-unity-ucs-list))
870 (defun mm-xemacs-find-mime-charset-1 (begin end)
871 "Determine which MIME charset to use to send region as message.
872 This uses the XEmacs-specific latin-unity package to better handle the
873 case where identical characters from diverse ISO-8859-? character sets
874 can be encoded using a single one of the corresponding coding systems.
876 It treats `mm-coding-system-priorities' as the list of preferred
877 coding systems; a useful example setting for this list in Western
878 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
879 to the very standard Latin 1 coding system, and only move to coding
880 systems that are less supported as is necessary to encode the
881 characters that exist in the buffer.
883 Latin Unity doesn't know about those non-ASCII Roman characters that
884 are available in various East Asian character sets. As such, its
885 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
886 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
887 But this is very much a corner case, so don't worry about it."
888 (let ((systems mm-coding-system-priorities) csets psets curset)
890 ;; Load the Latin Unity library, if available.
891 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
892 (require 'latin-unity))
894 ;; Now, can we use it?
895 (if (featurep 'latin-unity)
897 (setq csets (latin-unity-representations-feasible-region begin end)
898 psets (latin-unity-representations-present-region begin end))
902 ;; Pass back the first coding system in the preferred list
903 ;; that can encode the whole region.
904 (dolist (curset systems)
905 (setq curset (latin-unity-massage-name 'buffer-default curset))
907 ;; If the coding system is a universal coding system, then
908 ;; it can certainly encode all the characters in the region.
909 (if (memq curset latin-unity-ucs-list)
910 (throw 'done (list curset)))
912 ;; If a coding system isn't universal, and isn't in
913 ;; the list that latin unity knows about, we can't
914 ;; decide whether to use it here. Leave that until later
915 ;; in `mm-find-mime-charset-region' function, whence we
917 (unless (memq curset latin-unity-coding-systems)
920 ;; Right, we know about this coding system, and it may
921 ;; conceivably be able to encode all the characters in
923 (if (latin-unity-maybe-remap begin end curset csets psets t)
924 (throw 'done (list curset))))
926 ;; Can't encode using anything from the
927 ;; `mm-coding-system-priorities' list.
928 ;; Leave `mm-find-mime-charset' to do most of the work.
931 ;; Right, latin unity isn't available; let `mm-find-charset-region'
932 ;; take its default action, which equally applies to GNU Emacs.
935 (defmacro mm-xemacs-find-mime-charset (begin end)
936 (when (featurep 'xemacs)
937 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
939 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
940 "Return the MIME charsets needed to encode the region between B and E.
941 nil means ASCII, a single-element list represents an appropriate MIME
942 charset, and a longer list means no appropriate charset."
944 ;; The return possibilities of this function are a mess...
945 (or (and (mm-multibyte-p)
946 mm-use-find-coding-systems-region
947 ;; Find the mime-charset of the most preferred coding
948 ;; system that has one.
949 (let ((systems (find-coding-systems-region b e)))
950 (when mm-coding-system-priorities
952 (sort systems 'mm-sort-coding-systems-predicate)))
953 (setq systems (delq 'compound-text systems))
954 (unless (equal systems '(undecided))
956 (let* ((head (pop systems))
957 (cs (or (coding-system-get head :mime-charset)
958 (coding-system-get head 'mime-charset))))
959 ;; The mime-charset (`x-ctext') of
960 ;; `compound-text' is not in the IANA list. We
961 ;; shouldn't normally use anything here with a
962 ;; mime-charset having an `x-' prefix.
963 ;; Fixme: Allow this to be overridden, since
964 ;; there is existing use of x-ctext.
965 ;; Also people apparently need the coding system
966 ;; `iso-2022-jp-3' (which Mule-UCS defines with
967 ;; mime-charset, though it's not valid).
969 (not (string-match "^[Xx]-" (symbol-name cs)))
970 ;; UTF-16 of any variety is invalid for
971 ;; text parts and, unfortunately, has
972 ;; mime-charset defined both in Mule-UCS
973 ;; and versions of Emacs. (The name
974 ;; might be `mule-utf-16...' or
976 (not (string-match "utf-16" (symbol-name cs))))
978 charsets (list cs))))))
980 ;; If we're XEmacs, and some coding system is appropriate,
981 ;; mm-xemacs-find-mime-charset will return an appropriate list.
982 ;; Otherwise, we'll get nil, and the next setq will get invoked.
983 (setq charsets (mm-xemacs-find-mime-charset b e))
985 ;; We're not multibyte, or a single coding system won't cover it.
987 (mm-delete-duplicates
988 (mapcar 'mm-mime-charset
990 (mm-find-charset-region b e))))))
991 (if (and (> (length charsets) 1)
992 (memq 'iso-8859-15 charsets)
993 (memq 'iso-8859-15 hack-charsets)
994 (save-excursion (mm-iso-8859-x-to-15-region b e)))
995 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
996 mm-iso-8859-15-compatible))
997 (if (and (memq 'iso-2022-jp-2 charsets)
998 (memq 'iso-2022-jp-2 hack-charsets))
999 (setq charsets (delq 'iso-2022-jp charsets)))
1000 ;; Attempt to reduce the number of charsets if utf-8 is available.
1001 (if (and (featurep 'xemacs)
1002 (> (length charsets) 1)
1003 (mm-coding-system-p 'utf-8))
1004 (let ((mm-coding-system-priorities
1005 (cons 'utf-8 mm-coding-system-priorities)))
1007 (mm-delete-duplicates
1008 (mapcar 'mm-mime-charset
1010 (mm-find-charset-region b e)))))))
1013 (defmacro mm-with-unibyte-buffer (&rest forms)
1014 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1015 Use unibyte mode for this."
1016 `(let (default-enable-multibyte-characters)
1017 (with-temp-buffer ,@forms)))
1018 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
1019 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
1021 (defmacro mm-with-multibyte-buffer (&rest forms)
1022 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1023 Use multibyte mode for this."
1024 `(let ((default-enable-multibyte-characters t))
1025 (with-temp-buffer ,@forms)))
1026 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
1027 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
1029 (defmacro mm-with-unibyte-current-buffer (&rest forms)
1030 "Evaluate FORMS with current buffer temporarily made unibyte.
1031 Also bind `default-enable-multibyte-characters' to nil.
1032 Equivalent to `progn' in XEmacs
1034 NOTE: Use this macro with caution in multibyte buffers (it is not
1035 worth using this macro in unibyte buffers of course). Use of
1036 `(set-buffer-multibyte t)', which is run finally, is generally
1037 harmful since it is likely to modify existing data in the buffer.
1038 For instance, it converts \"\\300\\255\" into \"\\255\" in
1039 Emacs 23 (unicode)."
1040 (let ((multibyte (make-symbol "multibyte"))
1041 (buffer (make-symbol "buffer")))
1043 (let ((,multibyte enable-multibyte-characters)
1044 (,buffer (current-buffer)))
1046 (let (default-enable-multibyte-characters)
1047 (set-buffer-multibyte nil)
1049 (set-buffer ,buffer)
1050 (set-buffer-multibyte ,multibyte)))
1051 (let (default-enable-multibyte-characters)
1053 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
1054 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
1056 (defmacro mm-with-unibyte (&rest forms)
1057 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
1058 `(let (default-enable-multibyte-characters)
1060 (put 'mm-with-unibyte 'lisp-indent-function 0)
1061 (put 'mm-with-unibyte 'edebug-form-spec '(body))
1063 (defmacro mm-with-multibyte (&rest forms)
1064 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
1065 `(let ((default-enable-multibyte-characters t))
1067 (put 'mm-with-multibyte 'lisp-indent-function 0)
1068 (put 'mm-with-multibyte 'edebug-form-spec '(body))
1070 (defun mm-find-charset-region (b e)
1071 "Return a list of Emacs charsets in the region B to E."
1073 ((and (mm-multibyte-p)
1074 (fboundp 'find-charset-region))
1075 ;; Remove composition since the base charsets have been included.
1076 ;; Remove eight-bit-*, treat them as ascii.
1077 (let ((css (find-charset-region b e)))
1078 (mapcar (lambda (cs) (setq css (delq cs css)))
1079 '(composition eight-bit-control eight-bit-graphic
1083 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
1086 (narrow-to-region b e)
1087 (goto-char (point-min))
1088 (skip-chars-forward "\0-\177")
1093 (and (boundp 'current-language-environment)
1094 (car (last (assq 'charset
1095 (assoc current-language-environment
1096 language-info-alist))))))
1097 (if (eq charset 'ascii) (setq charset nil))
1100 (car (last (assq mail-parse-charset
1101 mm-mime-mule-charset-alist)))))
1102 (list 'ascii (or charset 'latin-iso8859-1)))))))))
1104 (defun mm-auto-mode-alist ()
1105 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
1106 (let ((alist auto-mode-alist)
1109 (when (listp (cdar alist))
1110 (push (car alist) out))
1114 (defvar mm-inhibit-file-name-handlers
1115 '(jka-compr-handler image-file-handler epa-file-handler)
1116 "A list of handlers doing (un)compression (etc) thingies.")
1118 (defun mm-insert-file-contents (filename &optional visit beg end replace
1120 "Like `insert-file-contents', but only reads in the file.
1121 A buffer may be modified in several ways after reading into the buffer due
1122 to advanced Emacs features, such as file-name-handlers, format decoding,
1123 `find-file-hooks', etc.
1124 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
1125 This function ensures that none of these modifications will take place."
1126 (let* ((format-alist nil)
1127 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
1128 (default-major-mode 'fundamental-mode)
1129 (enable-local-variables nil)
1130 (after-insert-file-functions nil)
1131 (enable-local-eval nil)
1132 (inhibit-file-name-operation (if inhibit
1133 'insert-file-contents
1134 inhibit-file-name-operation))
1135 (inhibit-file-name-handlers
1137 (append mm-inhibit-file-name-handlers
1138 inhibit-file-name-handlers)
1139 inhibit-file-name-handlers))
1140 (ffh (if (boundp 'find-file-hook)
1143 (val (symbol-value ffh)))
1146 (insert-file-contents filename visit beg end replace)
1149 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1150 "Append the contents of the region to the end of file FILENAME.
1151 When called from a function, expects three arguments,
1152 START, END and FILENAME. START and END are buffer positions
1153 saying what text to write.
1154 Optional fourth argument specifies the coding system to use when
1156 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1157 (let ((coding-system-for-write
1158 (or codesys mm-text-coding-system-for-write
1159 mm-text-coding-system))
1160 (inhibit-file-name-operation (if inhibit
1162 inhibit-file-name-operation))
1163 (inhibit-file-name-handlers
1165 (append mm-inhibit-file-name-handlers
1166 inhibit-file-name-handlers)
1167 inhibit-file-name-handlers)))
1168 (write-region start end filename t 'no-message)
1169 (message "Appended to %s" filename)))
1171 (defun mm-write-region (start end filename &optional append visit lockname
1172 coding-system inhibit)
1174 "Like `write-region'.
1175 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1176 (let ((coding-system-for-write
1177 (or coding-system mm-text-coding-system-for-write
1178 mm-text-coding-system))
1179 (inhibit-file-name-operation (if inhibit
1181 inhibit-file-name-operation))
1182 (inhibit-file-name-handlers
1184 (append mm-inhibit-file-name-handlers
1185 inhibit-file-name-handlers)
1186 inhibit-file-name-handlers)))
1187 (write-region start end filename append visit lockname)))
1189 ;; It is not a MIME function, but some MIME functions use it.
1190 (if (and (fboundp 'make-temp-file)
1192 (let ((def (symbol-function 'make-temp-file)))
1193 (and (byte-code-function-p def)
1194 (setq def (if (fboundp 'compiled-function-arglist)
1196 (eval (list 'compiled-function-arglist def))
1199 (eq (nth 3 def) 'suffix)))))
1200 (defalias 'mm-make-temp-file 'make-temp-file)
1201 ;; Stolen (and modified for XEmacs) from Emacs 22.
1202 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1203 "Create a temporary file.
1204 The returned file name (created by appending some random characters at the end
1205 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1206 is guaranteed to point to a newly created empty file.
1207 You can then use `write-region' to write new data into the file.
1209 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1211 If SUFFIX is non-nil, add that at the end of the file name."
1212 (let ((umask (default-file-modes))
1216 ;; Create temp files with strict access rights. It's easy to
1217 ;; loosen them later, whereas it's impossible to close the
1218 ;; time-window of loose permissions otherwise.
1219 (set-default-file-modes 448)
1220 (while (condition-case err
1226 (if (fboundp 'temp-directory)
1229 temporary-file-directory))))
1231 (setq file (concat file suffix)))
1233 (make-directory file)
1234 ;; NOTE: This is unsafe if Emacs 20
1235 ;; users and XEmacs users don't use
1236 ;; a secure temp directory.
1237 (gmm-write-region "" nil file nil 'silent
1240 (file-already-exists t)
1241 ;; The XEmacs version of `make-directory' issues
1243 (file-error (or (and (featurep 'xemacs)
1244 (file-exists-p file))
1245 (signal (car err) (cdr err)))))
1246 ;; the file was somehow created by someone else between
1247 ;; `make-temp-name' and `write-region', let's try again.
1251 (set-default-file-modes umask)))))
1253 (defun mm-image-load-path (&optional package)
1255 (dolist (path load-path (nreverse result))
1258 (setq dir (concat (file-name-directory
1259 (directory-file-name path))
1260 "etc/images/" (or package "gnus/")))))
1262 (push path result))))
1264 ;; Fixme: This doesn't look useful where it's used.
1265 (if (fboundp 'detect-coding-region)
1266 (defun mm-detect-coding-region (start end)
1267 "Like `detect-coding-region' except returning the best one."
1268 (let ((coding-systems
1269 (detect-coding-region start end)))
1270 (or (car-safe coding-systems)
1272 (defun mm-detect-coding-region (start end)
1273 (let ((point (point)))
1275 (skip-chars-forward "\0-\177" end)
1277 (if (eq (point) end) 'ascii (mm-guess-charset))
1278 (goto-char point)))))
1280 (if (fboundp 'coding-system-get)
1281 (defun mm-detect-mime-charset-region (start end)
1282 "Detect MIME charset of the text in the region between START and END."
1283 (let ((cs (mm-detect-coding-region start end)))
1284 (or (coding-system-get cs :mime-charset)
1285 (coding-system-get cs 'mime-charset))))
1286 (defun mm-detect-mime-charset-region (start end)
1287 "Detect MIME charset of the text in the region between START and END."
1288 (let ((cs (mm-detect-coding-region start end)))
1292 (unless (fboundp 'coding-system-to-mime-charset)
1293 (defalias 'coding-system-to-mime-charset 'ignore)))
1295 (defun mm-coding-system-to-mime-charset (coding-system)
1296 "Return the MIME charset corresponding to CODING-SYSTEM.
1297 To make this function work with XEmacs, the APEL package is required."
1299 (or (and (fboundp 'coding-system-get)
1300 (or (coding-system-get coding-system :mime-charset)
1301 (coding-system-get coding-system 'mime-charset)))
1302 (and (featurep 'xemacs)
1303 (or (and (fboundp 'coding-system-to-mime-charset)
1304 (not (eq (symbol-function 'coding-system-to-mime-charset)
1306 (and (condition-case nil
1309 (fboundp 'coding-system-to-mime-charset)))
1310 (coding-system-to-mime-charset coding-system)))))
1313 (require 'jka-compr))
1315 (defun mm-decompress-buffer (filename &optional inplace force)
1316 "Decompress buffer's contents, depending on jka-compr.
1317 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1318 agrees with `jka-compr-compression-info-list', decompression is done.
1319 Signal an error if FORCE is neither nil nor t and compressed data are
1320 not decompressed because `auto-compression-mode' is disabled.
1321 If INPLACE is nil, return decompressed data or nil without modifying
1322 the buffer. Otherwise, replace the buffer's contents with the
1323 decompressed data. The buffer's multibyteness must be turned off."
1326 (prog1 t (require 'jka-compr))
1327 (and (fboundp 'jka-compr-installed-p)
1328 (jka-compr-installed-p))))
1329 (let ((info (jka-compr-get-compression-info filename)))
1331 (unless (or (memq force (list nil t))
1332 (jka-compr-installed-p))
1334 (let ((prog (jka-compr-info-uncompress-program info))
1335 (args (jka-compr-info-uncompress-args info))
1336 (msg (format "%s %s..."
1337 (jka-compr-info-uncompress-message info)
1339 (err-file (jka-compr-make-temp-name))
1340 (cur (current-buffer))
1341 (coding-system-for-read mm-binary-coding-system)
1342 (coding-system-for-write mm-binary-coding-system)
1345 (mm-with-unibyte-buffer
1346 (insert-buffer-substring cur)
1349 (unless (memq (apply 'call-process-region
1350 (point-min) (point-max)
1351 prog t (list t err-file) nil args)
1352 jka-compr-acceptable-retval-list)
1356 (delete "" (split-string
1358 (insert-file-contents err-file)
1364 (format "Error while executing \"%s %s < %s\""
1365 prog (mapconcat 'identity args " ")
1367 (setq retval (buffer-string)))
1369 (setq err-msg (error-message-string err)))))
1370 (when (file-exists-p err-file)
1371 (ignore-errors (jka-compr-delete-temp-file err-file)))
1374 (delete-region (point-min) (point-max))
1377 (message "%s" (or err-msg (concat msg "done")))
1381 (unless (fboundp 'coding-system-name)
1382 (defalias 'coding-system-name 'ignore))
1383 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1384 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1385 (unless (fboundp 'find-operation-coding-system)
1386 (defalias 'find-operation-coding-system 'ignore)))
1388 (defun mm-find-buffer-file-coding-system (&optional filename)
1389 "Find coding system used to decode the contents of the current buffer.
1390 This function looks for the coding system magic cookie or examines the
1391 coding system specified by `file-coding-system-alist' being associated
1392 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1393 gzip, bzip2, etc. are allowed."
1395 (setq filename buffer-file-name))
1397 (let ((decomp (unless ;; No worth to examine charset of tar files.
1400 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1402 (mm-decompress-buffer filename nil t))))
1404 (set-buffer (let (default-enable-multibyte-characters)
1405 (generate-new-buffer " *temp*")))
1407 (setq filename (file-name-sans-extension filename)))
1408 (goto-char (point-min))
1411 ((boundp 'set-auto-coding-function) ;; Emacs
1413 (or (funcall (symbol-value 'set-auto-coding-function)
1414 filename (- (point-max) (point-min)))
1415 (car (find-operation-coding-system 'insert-file-contents
1417 (let (auto-coding-alist)
1419 (funcall (symbol-value 'set-auto-coding-function)
1420 nil (- (point-max) (point-min)))
1422 ((featurep 'file-coding) ;; XEmacs
1423 (let ((case-fold-search t)
1424 (end (point-at-eol))
1427 (and (re-search-forward "-\\*-+[\t ]*" end t)
1429 (setq start (match-end 0))
1430 (re-search-forward "[\t ]*-+\\*-" end t))
1432 (setq end (match-beginning 0))
1434 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1436 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1438 (find-coding-system (setq codesys
1439 (intern (match-string 1))))
1441 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1444 (setq start (match-end 0))
1445 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1447 (setq end (match-beginning 0))
1450 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1452 (find-coding-system (setq codesys
1453 (intern (match-string 1))))
1456 (goto-char (point-min))
1457 (setq case-fold-search nil)
1458 (re-search-forward "^;;;coding system: "
1459 ;;(+ (point-min) 3000) t))
1461 (looking-at "[^\t\n\r ]+")
1463 (setq codesys (intern (match-string 0))))
1467 (find-file-coding-system-for-read-from-filename
1469 (coding-system-name (coding-system-base codesys)))))))
1471 (kill-buffer (current-buffer)))))))
1475 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1476 ;;; mm-util.el ends here