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, 2008 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 3 of the License, or
13 ;; (at your option) any later version.
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. If not, see <http://www.gnu.org/licenses/>.
29 (unless (fboundp 'declare-function) (defmacro declare-function (&rest r))))
31 (eval-when-compile (require 'cl))
35 (if (featurep 'xemacs)
36 (unless (ignore-errors
37 (require 'timer-funcs))
41 (defvar mm-mime-mule-charset-alist )
46 (let ((nfunc (intern (format "mm-%s" (car elem)))))
47 (if (fboundp (car elem))
48 (defalias nfunc (car elem))
49 (defalias nfunc (cdr elem)))))
50 '((coding-system-list . 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)
106 (set-buffer-multibyte . ignore)
108 . (lambda (buffer-name)
109 "Returns non-nil if a buffer named BUFFER-NAME gets a special frame."
110 (and special-display-function
111 (or (and (member buffer-name special-display-buffer-names) t)
112 (cdr (assoc buffer-name special-display-buffer-names))
114 (dolist (elem special-display-regexps)
116 (string-match elem buffer-name)
120 (string-match (car elem) buffer-name)
121 (throw 'return (cdr elem))))))))))))
124 (if (featurep 'xemacs)
125 (if (featurep 'file-coding)
126 ;; Don't modify string if CODING-SYSTEM is nil.
128 (defun mm-decode-coding-string (str coding-system)
130 (decode-coding-string str coding-system)
132 (defun mm-encode-coding-string (str coding-system)
134 (encode-coding-string str coding-system)
136 (defun mm-decode-coding-region (start end coding-system)
138 (decode-coding-region start end coding-system)))
139 (defun mm-encode-coding-region (start end coding-system)
141 (encode-coding-region start end coding-system))))
142 (defun mm-decode-coding-string (str coding-system) str)
143 (defun mm-encode-coding-string (str coding-system) str)
144 (defalias 'mm-decode-coding-region 'ignore)
145 (defalias 'mm-encode-coding-region 'ignore))
146 (defalias 'mm-decode-coding-string 'decode-coding-string)
147 (defalias 'mm-encode-coding-string 'encode-coding-string)
148 (defalias 'mm-decode-coding-region 'decode-coding-region)
149 (defalias 'mm-encode-coding-region 'encode-coding-region)))
151 (defalias 'mm-string-to-multibyte
155 ((fboundp 'string-to-multibyte)
156 'string-to-multibyte)
159 "Return a multibyte string with the same individual chars as string."
161 (lambda (ch) (mm-string-as-multibyte (char-to-string ch)))
165 (defalias 'mm-char-or-char-int-p
167 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
168 ((fboundp 'char-valid-p) 'char-valid-p)
171 ;; Fixme: This seems always to be used to read a MIME charset, so it
172 ;; should be re-named and fixed (in Emacs) to offer completion only on
173 ;; proper charset names (base coding systems which have a
174 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
176 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
177 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
178 ;; Actually, there should be an `mm-coding-system-mime-charset'.
180 (defalias 'mm-read-coding-system
182 ((fboundp 'read-coding-system)
183 (if (and (featurep 'xemacs)
184 (<= (string-to-number emacs-version) 21.1))
185 (lambda (prompt &optional default-coding-system)
186 (read-coding-system prompt))
187 'read-coding-system))
188 (t (lambda (prompt &optional default-coding-system)
189 "Prompt the user for a coding system."
191 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
192 mm-mime-mule-charset-alist)))))))
194 (defvar mm-coding-system-list nil)
195 (defun mm-get-coding-system-list ()
196 "Get the coding system list."
197 (or mm-coding-system-list
198 (setq mm-coding-system-list (mm-coding-system-list))))
200 (defun mm-coding-system-p (cs)
201 "Return non-nil if CS is a symbol naming a coding system.
202 In XEmacs, also return non-nil if CS is a coding system object.
203 If CS is available, return CS itself in Emacs, and return a coding
204 system object in XEmacs."
205 (if (fboundp 'find-coding-system)
206 (and cs (find-coding-system cs))
207 (if (fboundp 'coding-system-p)
208 (when (coding-system-p cs)
211 (car (memq cs (mm-get-coding-system-list))))))
213 (defun mm-codepage-setup (number &optional alias)
214 "Create a coding system cpNUMBER.
215 The coding system is created using `codepage-setup'. If ALIAS is
216 non-nil, an alias is created and added to
217 `mm-charset-synonym-alist'. If ALIAS is a string, it's used as
218 the alias. Else windows-NUMBER is used."
220 (let ((completion-ignore-case t)
221 (candidates (if (fboundp 'cp-supported-codepages)
222 (cp-supported-codepages)
223 ;; Removed in Emacs 23 (unicode), so signal an error:
224 (error "`codepage-setup' not present in this Emacs version."))))
225 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
226 nil t nil nil "437"))))
228 (setq alias (if (stringp alias)
230 (intern (format "windows-%s" number)))))
231 (let* ((cp (intern (format "cp%s" number))))
232 (unless (mm-coding-system-p cp)
233 (if (fboundp 'codepage-setup) ; silence compiler
234 (codepage-setup number)
235 (error "`codepage-setup' not present in this Emacs version.")))
237 ;; Don't add alias if setup of cp failed.
238 (mm-coding-system-p cp))
239 (add-to-list 'mm-charset-synonym-alist (cons alias cp)))))
241 (defvar mm-charset-synonym-alist
243 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
244 ,@(unless (mm-coding-system-p 'x-ctext)
245 '((x-ctext . ctext)))
246 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_ in 8
248 ,@(unless (mm-coding-system-p 'iso-8859-15)
249 '((iso-8859-15 . iso-8859-1)))
250 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
251 ,@(unless (mm-coding-system-p 'big5-hkscs)
252 '((big5-hkscs . big5)))
253 ;; A Microsoft misunderstanding.
254 ,@(when (and (not (mm-coding-system-p 'unicode))
255 (mm-coding-system-p 'utf-16-le))
256 '((unicode . utf-16-le)))
257 ;; A Microsoft misunderstanding.
258 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
259 (if (mm-coding-system-p 'cp949)
260 '((ks_c_5601-1987 . cp949))
261 '((ks_c_5601-1987 . euc-kr))))
262 ;; Windows-31J is Windows Codepage 932.
263 ,@(when (and (not (mm-coding-system-p 'windows-31j))
264 (mm-coding-system-p 'cp932))
265 '((windows-31j . cp932)))
266 ;; Charset name: GBK, Charset aliases: CP936, MS936, windows-936
267 ;; http://www.iana.org/assignments/charset-reg/GBK
268 ;; Emacs 22.1 has cp936, but not gbk, so we alias it:
269 ,@(when (and (not (mm-coding-system-p 'gbk))
270 (mm-coding-system-p 'cp936))
272 ;; UTF8 is a bogus name for UTF-8
273 ,@(when (and (not (mm-coding-system-p 'utf8))
274 (mm-coding-system-p 'utf-8))
276 ;; ISO8859-1 is a bogus name for ISO-8859-1
277 ,@(when (and (not (mm-coding-system-p 'iso8859-1))
278 (mm-coding-system-p 'iso-8859-1))
279 '((iso8859-1 . iso-8859-1)))
280 ;; ISO_8859-1 is a bogus name for ISO-8859-1
281 ,@(when (and (not (mm-coding-system-p 'iso_8859-1))
282 (mm-coding-system-p 'iso-8859-1))
283 '((iso_8859-1 . iso-8859-1)))
285 "A mapping from unknown or invalid charset names to the real charset names.
287 See `mm-codepage-iso-8859-list' and `mm-codepage-ibm-list'.")
289 (defcustom mm-codepage-iso-8859-list
290 (list 1250 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
291 ;; Outlook users in Czech republic. Use this to allow reading of
292 ;; their e-mails. cp1250 should be defined by M-x codepage-setup
294 '(1252 . 1) ;; Windows-1252 is a superset of iso-8859-1 (West
295 ;; Europe). See also `gnus-article-dumbquotes-map'.
296 '(1254 . 9) ;; Windows-1254 is a superset of iso-8859-9 (Turkish).
297 '(1255 . 8));; Windows-1255 is a superset of iso-8859-8 (Hebrew).
298 "A list of Windows codepage numbers and iso-8859 charset numbers.
300 If an element is a number corresponding to a supported windows
301 codepage, appropriate entries to `mm-charset-synonym-alist' are
302 added by `mm-setup-codepage-iso-8859'. An element may also be a
303 cons cell where the car is a codepage number and the cdr is the
304 corresponding number of an iso-8859 charset."
305 :type '(list (set :inline t
306 (const 1250 :tag "Central and East European")
307 (const (1252 . 1) :tag "West European")
308 (const (1254 . 9) :tag "Turkish")
309 (const (1255 . 8) :tag "Hebrew"))
313 (integer :tag "Windows codepage number")
314 (cons (integer :tag "Windows codepage number")
315 (integer :tag "iso-8859 charset number")))))
316 :version "22.1" ;; Gnus 5.10.9
319 (defcustom mm-codepage-ibm-list
320 (list 437 ;; (US etc.)
324 863 ;; (Canadian French)
329 866 ;; (Cyrillic - Russian)
334 ;; In Emacs 23 (unicode), cp... and ibm... are aliases.
335 ;; Cf. http://thread.gmane.org/v9lkng5nwy.fsf@marauder.physik.uni-ulm.de
336 "List of IBM codepage numbers.
338 The codepage mappings slighly differ between IBM and other vendors.
339 See \"ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/IBM/README.TXT\".
341 If an element is a number corresponding to a supported windows
342 codepage, appropriate entries to `mm-charset-synonym-alist' are
343 added by `mm-setup-codepage-ibm'."
344 :type '(list (set :inline t
345 (const 437 :tag "US etc.")
346 (const 860 :tag "Portugal")
347 (const 861 :tag "Iceland")
348 (const 862 :tag "Israel")
349 (const 863 :tag "Canadian French")
350 (const 865 :tag "Nordic")
352 (const 850 :tag "Latin 1")
353 (const 855 :tag "Cyrillic")
354 (const 866 :tag "Cyrillic - Russian")
355 (const 857 :tag "Turkish")
356 (const 864 :tag "Arabic")
357 (const 869 :tag "Greek")
358 (const 874 :tag "Thai"))
361 (integer :tag "Codepage number")))
362 :version "22.1" ;; Gnus 5.10.9
365 (defun mm-setup-codepage-iso-8859 (&optional list)
366 "Add appropriate entries to `mm-charset-synonym-alist'.
367 Unless LIST is given, `mm-codepage-iso-8859-list' is used."
369 (setq list mm-codepage-iso-8859-list))
371 (let (cp windows iso)
373 (setq cp (intern (format "cp%d" (car i)))
374 windows (intern (format "windows-%d" (car i)))
375 iso (intern (format "iso-8859-%d" (cdr i))))
376 (setq cp (intern (format "cp%d" i))
377 windows (intern (format "windows-%d" i))))
378 (unless (mm-coding-system-p windows)
379 (if (mm-coding-system-p cp)
380 (add-to-list 'mm-charset-synonym-alist (cons windows cp))
381 (add-to-list 'mm-charset-synonym-alist (cons windows iso)))))))
383 (defun mm-setup-codepage-ibm (&optional list)
384 "Add appropriate entries to `mm-charset-synonym-alist'.
385 Unless LIST is given, `mm-codepage-ibm-list' is used."
387 (setq list mm-codepage-ibm-list))
388 (dolist (number list)
389 (let ((ibm (intern (format "ibm%d" number)))
390 (cp (intern (format "cp%d" number))))
391 (when (and (not (mm-coding-system-p ibm))
392 (mm-coding-system-p cp))
393 (add-to-list 'mm-charset-synonym-alist (cons ibm cp))))))
396 (mm-setup-codepage-iso-8859)
397 (mm-setup-codepage-ibm)
399 ;; Note: this has to be defined before `mm-charset-to-coding-system'.
400 (defcustom mm-charset-eval-alist
401 (if (featurep 'xemacs)
402 nil ;; I don't know what would be useful for XEmacs.
403 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
404 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
405 (windows-1250 . (mm-codepage-setup 1250 t))
406 (windows-1251 . (mm-codepage-setup 1251 t))
407 (windows-1253 . (mm-codepage-setup 1253 t))
408 (windows-1257 . (mm-codepage-setup 1257 t))))
409 "An alist of (CHARSET . FORM) pairs.
410 If an article is encoded in an unknown CHARSET, FORM is
411 evaluated. This allows to load additional libraries providing
412 charsets on demand. If supported by your Emacs version, you
413 could use `autoload-coding-system' here."
414 :version "22.1" ;; Gnus 5.10.9
415 :type '(list (set :inline t
416 (const (windows-1250 . (mm-codepage-setup 1250 t)))
417 (const (windows-1251 . (mm-codepage-setup 1251 t)))
418 (const (windows-1253 . (mm-codepage-setup 1253 t)))
419 (const (windows-1257 . (mm-codepage-setup 1257 t)))
420 (const (cp850 . (mm-codepage-setup 850 nil))))
423 (cons (symbol :tag "charset")
424 (symbol :tag "form"))))
426 (put 'mm-charset-eval-alist 'risky-local-variable t)
428 ;; Note: this function has to be defined before `mm-charset-override-alist'
429 ;; since it will use this function in order to determine its default value
430 ;; when loading mm-util.elc.
431 (defun mm-charset-to-coding-system (charset &optional lbt
432 allow-override silent)
433 "Return coding-system corresponding to CHARSET.
434 CHARSET is a symbol naming a MIME charset.
435 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
436 used as the line break code type of the coding system.
438 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
439 map undesired charset names to their replacement. This should
440 only be used for decoding, not for encoding.
442 A non-nil value of SILENT means don't issue a warning even if CHARSET
444 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
445 (when (stringp charset)
446 (setq charset (intern (downcase charset))))
448 (setq charset (intern (format "%s-%s" charset lbt))))
452 ;; Running in a non-MULE environment.
453 ((or (null (mm-get-coding-system-list))
454 (not (fboundp 'coding-system-get)))
456 ;; Check override list quite early. Should only used for decoding, not for
459 (let ((cs (cdr (assq charset mm-charset-override-alist))))
460 (and cs (mm-coding-system-p cs) cs))))
462 ((eq charset 'us-ascii)
464 ;; Check to see whether we can handle this charset. (This depends
465 ;; on there being some coding system matching each `mime-charset'
466 ;; property defined, as there should be.)
467 ((and (mm-coding-system-p charset)
468 ;;; Doing this would potentially weed out incorrect charsets.
470 ;;; (eq charset (coding-system-get charset 'mime-charset))
473 ;; Eval expressions from `mm-charset-eval-alist'
474 ((let* ((el (assq charset mm-charset-eval-alist))
481 (condition-case nil (eval form) (error nil))
482 ;; (message "Failed to eval `%s'" form))
483 (mm-coding-system-p cs)
484 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
486 ;; Translate invalid charsets.
487 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
489 (mm-coding-system-p cs)
491 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
494 ;; Last resort: search the coding system list for entries which
495 ;; have the right mime-charset in case the canonical name isn't
496 ;; defined (though it should be).
498 ;; mm-get-coding-system-list returns a list of cs without lbt.
500 (dolist (c (mm-get-coding-system-list))
502 (eq charset (or (coding-system-get c :mime-charset)
503 (coding-system-get c 'mime-charset))))
505 (unless (or silent cs)
506 ;; Warn the user about unknown charset:
507 (if (fboundp 'gnus-message)
508 (gnus-message 7 "Unknown charset: %s" charset)
509 (message "Unknown charset: %s" charset)))
512 ;; Note: `mm-charset-to-coding-system' has to be defined before this.
513 (defcustom mm-charset-override-alist
514 ;; Note: pairs that cannot be used in the Emacs version currently running
517 (iso-8859-1 . windows-1252)
518 (iso-8859-8 . windows-1255)
519 (iso-8859-9 . windows-1254))
520 "A mapping from undesired charset names to their replacement.
522 You may add pairs like (iso-8859-1 . windows-1252) here,
523 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
524 superset of iso-8859-1."
531 (mapcar (lambda (pair)
532 (if (mm-charset-to-coding-system (cdr pair)
536 (iso-8859-1 . windows-1252)
537 (iso-8859-8 . windows-1255)
538 (iso-8859-9 . windows-1254)
539 (undecided . windows-1252)))))
540 (val (copy-sequence (default-value 'mm-charset-override-alist)))
543 (push (if (and (prog1
544 (setq pair (assq (caar val) defaults))
545 (setq defaults (delq pair defaults)))
546 (equal (car val) pair))
549 (const :format "(%v" ,(caar val))
550 (symbol :size 3 :format " . %v)\n" ,(cdar val))))
552 (setq val (cdr val)))
554 (push `(const ,(pop defaults)) rest))
557 `(set :inline t :format "%v" ,@(nreverse rest))
558 `(repeat :inline t :tag "Other options"
560 (symbol :size 3 :format "(%v")
561 (symbol :size 3 :format " . %v)\n")))))))
562 ;; Remove pairs that cannot be used in the Emacs version currently
563 ;; running. Note that this section will be evaluated when loading
565 :set (lambda (symbol value)
568 (mapcar (lambda (pair)
569 (if (mm-charset-to-coding-system (cdr pair)
573 :version "22.1" ;; Gnus 5.10.9
576 (defvar mm-binary-coding-system
578 ((mm-coding-system-p 'binary) 'binary)
579 ((mm-coding-system-p 'no-conversion) 'no-conversion)
581 "100% binary coding system.")
583 (defvar mm-text-coding-system
584 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
585 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
586 (and (mm-coding-system-p 'raw-text) 'raw-text))
587 mm-binary-coding-system)
588 "Text-safe coding system (For removing ^M).")
590 (defvar mm-text-coding-system-for-write nil
591 "Text coding system for write.")
593 (defvar mm-auto-save-coding-system
595 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
596 (if (memq system-type '(windows-nt ms-dos ms-windows))
597 (if (mm-coding-system-p 'utf-8-emacs-dos)
598 'utf-8-emacs-dos mm-binary-coding-system)
600 ((mm-coding-system-p 'emacs-mule)
601 (if (memq system-type '(windows-nt ms-dos ms-windows))
602 (if (mm-coding-system-p 'emacs-mule-dos)
603 'emacs-mule-dos mm-binary-coding-system)
605 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
606 (t mm-binary-coding-system))
607 "Coding system of auto save file.")
609 (defvar mm-universal-coding-system mm-auto-save-coding-system
610 "The universal coding system.")
612 ;; Fixme: some of the cars here aren't valid MIME charsets. That
613 ;; should only matter with XEmacs, though.
614 (defvar mm-mime-mule-charset-alist
616 (iso-8859-1 latin-iso8859-1)
617 (iso-8859-2 latin-iso8859-2)
618 (iso-8859-3 latin-iso8859-3)
619 (iso-8859-4 latin-iso8859-4)
620 (iso-8859-5 cyrillic-iso8859-5)
621 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
622 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
623 ;; charset is koi8-r, not iso-8859-5.
624 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
625 (iso-8859-6 arabic-iso8859-6)
626 (iso-8859-7 greek-iso8859-7)
627 (iso-8859-8 hebrew-iso8859-8)
628 (iso-8859-9 latin-iso8859-9)
629 (iso-8859-14 latin-iso8859-14)
630 (iso-8859-15 latin-iso8859-15)
631 (viscii vietnamese-viscii-lower)
632 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
633 (euc-kr korean-ksc5601)
634 (gb2312 chinese-gb2312)
636 (gb18030 gb18030-2-byte
637 gb18030-4-byte-bmp gb18030-4-byte-smp
638 gb18030-4-byte-ext-1 gb18030-4-byte-ext-2)
639 (big5 chinese-big5-1 chinese-big5-2)
641 (thai-tis620 thai-tis620)
642 (windows-1251 cyrillic-iso8859-5)
643 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
644 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
645 latin-jisx0201 japanese-jisx0208-1978
646 chinese-gb2312 japanese-jisx0208
647 korean-ksc5601 japanese-jisx0212)
648 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
649 latin-jisx0201 japanese-jisx0208-1978
650 chinese-gb2312 japanese-jisx0208
651 korean-ksc5601 japanese-jisx0212
652 chinese-cns11643-1 chinese-cns11643-2)
653 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
654 cyrillic-iso8859-5 greek-iso8859-7
655 latin-jisx0201 japanese-jisx0208-1978
656 chinese-gb2312 japanese-jisx0208
657 korean-ksc5601 japanese-jisx0212
658 chinese-cns11643-1 chinese-cns11643-2
659 chinese-cns11643-3 chinese-cns11643-4
660 chinese-cns11643-5 chinese-cns11643-6
662 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
663 japanese-jisx0213-1 japanese-jisx0213-2)
664 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
665 ,(cond ((fboundp 'unicode-precedence-list)
666 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
667 (unicode-precedence-list)))))
668 ((or (not (fboundp 'charsetp)) ;; non-Mule case
669 (charsetp 'unicode-a)
670 (not (mm-coding-system-p 'mule-utf-8)))
671 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
672 (t ;; If we have utf-8 we're in Mule 5+.
675 (coding-system-get 'mule-utf-8 'safe-charsets))))))
676 "Alist of MIME-charset/MULE-charsets.")
678 (defun mm-enrich-utf-8-by-mule-ucs ()
679 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
680 This function will run when the `un-define' module is loaded under
681 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
682 with Mule charsets. It is completely useless for Emacs."
683 (when (boundp 'unicode-basic-translation-charset-order-list)
689 'unicode-basic-translation-charset-order-list))))
690 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
693 (setq mm-mime-mule-charset-alist
694 (nconc mm-mime-mule-charset-alist
695 (list (cons 'utf-8 val))))))
698 ;; Correct by construction, but should be unnecessary for Emacs:
699 (if (featurep 'xemacs)
700 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
701 (when (and (fboundp 'coding-system-list)
702 (fboundp 'sort-coding-systems))
703 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
707 mime (or (coding-system-get cs :mime-charset); Emacs 23 (unicode)
708 (coding-system-get cs 'mime-charset)))
710 (not (eq t (setq mule
711 (coding-system-get cs 'safe-charsets))))
712 (not (assq mime alist)))
713 (push (cons mime (delq 'ascii mule)) alist)))
714 (setq mm-mime-mule-charset-alist (nreverse alist)))))
716 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
717 "A list of special charsets.
718 Valid elements include:
719 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
720 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
723 (defvar mm-iso-8859-15-compatible
724 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
725 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
726 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
728 (defvar mm-iso-8859-x-to-15-table
729 (and (fboundp 'coding-system-p)
730 (mm-coding-system-p 'iso-8859-15)
733 (if (mm-coding-system-p (car cs))
734 (let ((c (string-to-char
735 (decode-coding-string "\341" (car cs)))))
736 (cons (char-charset c)
739 (decode-coding-string "\341" 'iso-8859-15)) c)
740 (string-to-list (decode-coding-string (car (cdr cs))
743 mm-iso-8859-15-compatible))
744 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
746 (defcustom mm-coding-system-priorities
747 (if (boundp 'current-language-environment)
748 (let ((lang (symbol-value 'current-language-environment)))
749 (cond ((string= lang "Japanese")
750 ;; Japanese users prefer iso-2022-jp to euc-japan or
751 ;; shift_jis, however iso-8859-1 should be used when
752 ;; there are only ASCII text and Latin-1 characters.
753 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
754 "Preferred coding systems for encoding outgoing messages.
756 More than one suitable coding system may be found for some text.
757 By default, the coding system with the highest priority is used
758 to encode outgoing messages (see `sort-coding-systems'). If this
759 variable is set, it overrides the default priority."
761 :type '(repeat (symbol :tag "Coding system"))
765 (defvar mm-use-find-coding-systems-region
766 (fboundp 'find-coding-systems-region)
767 "Use `find-coding-systems-region' to find proper coding systems.
769 Setting it to nil is useful on Emacsen supporting Unicode if sending
770 mail with multiple parts is preferred to sending a Unicode one.")
772 ;;; Internal variables:
776 (defun mm-mule-charset-to-mime-charset (charset)
777 "Return the MIME charset corresponding to the given Mule CHARSET."
778 (if (and (fboundp 'find-coding-systems-for-charsets)
779 (fboundp 'sort-coding-systems))
780 (let ((css (sort (sort-coding-systems
781 (find-coding-systems-for-charsets (list charset)))
782 'mm-sort-coding-systems-predicate))
784 (while (and (not mime)
786 (when (setq cs (pop css))
787 (setq mime (or (coding-system-get cs :mime-charset)
788 (coding-system-get cs 'mime-charset)))))
790 (let ((alist (mapcar (lambda (cs)
791 (assq cs mm-mime-mule-charset-alist))
792 (sort (mapcar 'car mm-mime-mule-charset-alist)
793 'mm-sort-coding-systems-predicate)))
796 (when (memq charset (cdar alist))
797 (setq out (caar alist)
803 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
804 (boundp 'default-enable-multibyte-characters)
805 default-enable-multibyte-characters
806 (fboundp 'set-buffer-multibyte))
807 "True in Emacs with Mule.")
810 (defun mm-enable-multibyte ()
811 "Set the multibyte flag of the current buffer.
812 Only do this if the default value of `enable-multibyte-characters' is
813 non-nil. This is a no-op in XEmacs."
814 (set-buffer-multibyte 'to))
815 (defalias 'mm-enable-multibyte 'ignore))
818 (defun mm-disable-multibyte ()
819 "Unset the multibyte flag of in the current buffer.
820 This is a no-op in XEmacs."
821 (set-buffer-multibyte nil))
822 (defalias 'mm-disable-multibyte 'ignore)))
824 (defun mm-preferred-coding-system (charset)
825 ;; A typo in some Emacs versions.
826 (or (get-charset-property charset 'preferred-coding-system)
827 (get-charset-property charset 'prefered-coding-system)))
829 ;; Mule charsets shouldn't be used.
830 (defsubst mm-guess-charset ()
831 "Guess Mule charset from the language environment."
833 mail-parse-mule-charset ;; cached mule-charset
835 (setq mail-parse-mule-charset
836 (and (boundp 'current-language-environment)
839 (assoc current-language-environment
840 language-info-alist))))))
841 (if (or (not mail-parse-mule-charset)
842 (eq mail-parse-mule-charset 'ascii))
843 (setq mail-parse-mule-charset
844 (or (car (last (assq mail-parse-charset
845 mm-mime-mule-charset-alist)))
848 mail-parse-mule-charset)))
850 (defun mm-charset-after (&optional pos)
851 "Return charset of a character in current buffer at position POS.
852 If POS is nil, it defauls to the current point.
853 If POS is out of range, the value is nil.
854 If the charset is `composition', return the actual one."
855 (let ((char (char-after pos)) charset)
856 (if (< (mm-char-int char) 128)
857 (setq charset 'ascii)
858 ;; charset-after is fake in some Emacsen.
859 (setq charset (and (fboundp 'char-charset) (char-charset char)))
860 (if (eq charset 'composition) ; Mule 4
861 (let ((p (or pos (point))))
862 (cadr (find-charset-region p (1+ p))))
863 (if (and charset (not (memq charset '(ascii eight-bit-control
864 eight-bit-graphic))))
866 (mm-guess-charset))))))
868 (defun mm-mime-charset (charset)
869 "Return the MIME charset corresponding to the given Mule CHARSET."
870 (if (eq charset 'unknown)
871 (error "The message contains non-printable characters, please use attachment"))
872 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
873 ;; This exists in Emacs 20.
875 (and (mm-preferred-coding-system charset)
876 (or (coding-system-get
877 (mm-preferred-coding-system charset) :mime-charset)
879 (mm-preferred-coding-system charset) 'mime-charset)))
880 (and (eq charset 'ascii)
882 (mm-preferred-coding-system charset)
883 (mm-mule-charset-to-mime-charset charset))
884 ;; This is for XEmacs.
885 (mm-mule-charset-to-mime-charset charset)))
887 (if (fboundp 'delete-dups)
888 (defalias 'mm-delete-duplicates 'delete-dups)
889 (defun mm-delete-duplicates (list)
890 "Destructively remove `equal' duplicates from LIST.
891 Store the result in LIST and return it. LIST must be a proper list.
892 Of several `equal' occurrences of an element in LIST, the first
895 This is a compatibility function for Emacsen without `delete-dups'."
896 ;; Code from `subr.el' in Emacs 22:
899 (setcdr tail (delete (car tail) (cdr tail)))
900 (setq tail (cdr tail))))
903 ;; Fixme: This is used in places when it should be testing the
904 ;; default multibyteness. See mm-default-multibyte-p.
906 (if (and (not (featurep 'xemacs))
907 (boundp 'enable-multibyte-characters))
908 (defun mm-multibyte-p ()
909 "Non-nil if multibyte is enabled in the current buffer."
910 enable-multibyte-characters)
911 (defun mm-multibyte-p () (featurep 'mule))))
913 (defun mm-default-multibyte-p ()
914 "Return non-nil if the session is multibyte.
915 This affects whether coding conversion should be attempted generally."
917 (if (boundp 'default-enable-multibyte-characters)
918 default-enable-multibyte-characters
921 (defun mm-iso-8859-x-to-15-region (&optional b e)
922 (if (fboundp 'char-charset)
923 (let (charset item c inconvertible)
925 (if e (narrow-to-region b e))
926 (goto-char (point-min))
927 (skip-chars-forward "\0-\177")
930 ((not (setq item (assq (char-charset (setq c (char-after)))
931 mm-iso-8859-x-to-15-table)))
933 ((memq c (cdr (cdr item)))
934 (setq inconvertible t)
937 (insert-before-markers (prog1 (+ c (car (cdr item)))
939 (skip-chars-forward "\0-\177")))
940 (not inconvertible))))
942 (defun mm-sort-coding-systems-predicate (a b)
945 ;; Note: invalid entries are dropped silently
946 (and (setq cs (mm-coding-system-p cs))
947 (coding-system-base cs)))
948 mm-coding-system-priorities)))
949 (and (setq a (mm-coding-system-p a))
950 (if (setq b (mm-coding-system-p b))
951 (> (length (memq (coding-system-base a) priorities))
952 (length (memq (coding-system-base b) priorities)))
956 (autoload 'latin-unity-massage-name "latin-unity")
957 (autoload 'latin-unity-maybe-remap "latin-unity")
958 (autoload 'latin-unity-representations-feasible-region "latin-unity")
959 (autoload 'latin-unity-representations-present-region "latin-unity"))
961 (defvar latin-unity-coding-systems)
962 (defvar latin-unity-ucs-list)
964 (defun mm-xemacs-find-mime-charset-1 (begin end)
965 "Determine which MIME charset to use to send region as message.
966 This uses the XEmacs-specific latin-unity package to better handle the
967 case where identical characters from diverse ISO-8859-? character sets
968 can be encoded using a single one of the corresponding coding systems.
970 It treats `mm-coding-system-priorities' as the list of preferred
971 coding systems; a useful example setting for this list in Western
972 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
973 to the very standard Latin 1 coding system, and only move to coding
974 systems that are less supported as is necessary to encode the
975 characters that exist in the buffer.
977 Latin Unity doesn't know about those non-ASCII Roman characters that
978 are available in various East Asian character sets. As such, its
979 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
980 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
981 But this is very much a corner case, so don't worry about it."
982 (let ((systems mm-coding-system-priorities) csets psets curset)
984 ;; Load the Latin Unity library, if available.
985 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
986 (require 'latin-unity))
988 ;; Now, can we use it?
989 (if (featurep 'latin-unity)
991 (setq csets (latin-unity-representations-feasible-region begin end)
992 psets (latin-unity-representations-present-region begin end))
996 ;; Pass back the first coding system in the preferred list
997 ;; that can encode the whole region.
998 (dolist (curset systems)
999 (setq curset (latin-unity-massage-name 'buffer-default curset))
1001 ;; If the coding system is a universal coding system, then
1002 ;; it can certainly encode all the characters in the region.
1003 (if (memq curset latin-unity-ucs-list)
1004 (throw 'done (list curset)))
1006 ;; If a coding system isn't universal, and isn't in
1007 ;; the list that latin unity knows about, we can't
1008 ;; decide whether to use it here. Leave that until later
1009 ;; in `mm-find-mime-charset-region' function, whence we
1010 ;; have been called.
1011 (unless (memq curset latin-unity-coding-systems)
1014 ;; Right, we know about this coding system, and it may
1015 ;; conceivably be able to encode all the characters in
1017 (if (latin-unity-maybe-remap begin end curset csets psets t)
1018 (throw 'done (list curset))))
1020 ;; Can't encode using anything from the
1021 ;; `mm-coding-system-priorities' list.
1022 ;; Leave `mm-find-mime-charset' to do most of the work.
1025 ;; Right, latin unity isn't available; let `mm-find-charset-region'
1026 ;; take its default action, which equally applies to GNU Emacs.
1029 (defmacro mm-xemacs-find-mime-charset (begin end)
1030 (when (featurep 'xemacs)
1031 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
1033 (declare-function mm-delete-duplicates "mm-util" (list))
1035 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
1036 "Return the MIME charsets needed to encode the region between B and E.
1037 nil means ASCII, a single-element list represents an appropriate MIME
1038 charset, and a longer list means no appropriate charset."
1040 ;; The return possibilities of this function are a mess...
1041 (or (and (mm-multibyte-p)
1042 mm-use-find-coding-systems-region
1043 ;; Find the mime-charset of the most preferred coding
1044 ;; system that has one.
1045 (let ((systems (find-coding-systems-region b e)))
1046 (when mm-coding-system-priorities
1048 (sort systems 'mm-sort-coding-systems-predicate)))
1049 (setq systems (delq 'compound-text systems))
1050 (unless (equal systems '(undecided))
1052 (let* ((head (pop systems))
1053 (cs (or (coding-system-get head :mime-charset)
1054 (coding-system-get head 'mime-charset))))
1055 ;; The mime-charset (`x-ctext') of
1056 ;; `compound-text' is not in the IANA list. We
1057 ;; shouldn't normally use anything here with a
1058 ;; mime-charset having an `x-' prefix.
1059 ;; Fixme: Allow this to be overridden, since
1060 ;; there is existing use of x-ctext.
1061 ;; Also people apparently need the coding system
1062 ;; `iso-2022-jp-3' (which Mule-UCS defines with
1063 ;; mime-charset, though it's not valid).
1065 (not (string-match "^[Xx]-" (symbol-name cs)))
1066 ;; UTF-16 of any variety is invalid for
1067 ;; text parts and, unfortunately, has
1068 ;; mime-charset defined both in Mule-UCS
1069 ;; and versions of Emacs. (The name
1070 ;; might be `mule-utf-16...' or
1072 (not (string-match "utf-16" (symbol-name cs))))
1074 charsets (list cs))))))
1076 ;; If we're XEmacs, and some coding system is appropriate,
1077 ;; mm-xemacs-find-mime-charset will return an appropriate list.
1078 ;; Otherwise, we'll get nil, and the next setq will get invoked.
1079 (setq charsets (mm-xemacs-find-mime-charset b e))
1081 ;; Fixme: won't work for unibyte Emacs 23:
1083 ;; We're not multibyte, or a single coding system won't cover it.
1085 (mm-delete-duplicates
1086 (mapcar 'mm-mime-charset
1088 (mm-find-charset-region b e))))))
1089 (if (and (> (length charsets) 1)
1090 (memq 'iso-8859-15 charsets)
1091 (memq 'iso-8859-15 hack-charsets)
1092 (save-excursion (mm-iso-8859-x-to-15-region b e)))
1093 (dolist (x mm-iso-8859-15-compatible)
1094 (setq charsets (delq (car x) charsets))))
1095 (if (and (memq 'iso-2022-jp-2 charsets)
1096 (memq 'iso-2022-jp-2 hack-charsets))
1097 (setq charsets (delq 'iso-2022-jp charsets)))
1098 ;; Attempt to reduce the number of charsets if utf-8 is available.
1099 (if (and (featurep 'xemacs)
1100 (> (length charsets) 1)
1101 (mm-coding-system-p 'utf-8))
1102 (let ((mm-coding-system-priorities
1103 (cons 'utf-8 mm-coding-system-priorities)))
1105 (mm-delete-duplicates
1106 (mapcar 'mm-mime-charset
1108 (mm-find-charset-region b e)))))))
1111 (defmacro mm-with-unibyte-buffer (&rest forms)
1112 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1113 Use unibyte mode for this."
1115 (mm-disable-multibyte)
1117 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
1118 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
1120 (defmacro mm-with-multibyte-buffer (&rest forms)
1121 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1122 Use multibyte mode for this."
1124 (mm-enable-multibyte)
1126 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
1127 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
1129 (defmacro mm-with-unibyte-current-buffer (&rest forms)
1130 "Evaluate FORMS with current buffer temporarily made unibyte.
1131 Also bind `default-enable-multibyte-characters' to nil.
1132 Equivalent to `progn' in XEmacs
1134 NOTE: Use this macro with caution in multibyte buffers (it is not
1135 worth using this macro in unibyte buffers of course). Use of
1136 `(set-buffer-multibyte t)', which is run finally, is generally
1137 harmful since it is likely to modify existing data in the buffer.
1138 For instance, it converts \"\\300\\255\" into \"\\255\" in
1139 Emacs 23 (unicode)."
1140 (let ((multibyte (make-symbol "multibyte"))
1141 (buffer (make-symbol "buffer")))
1143 (let ((,multibyte enable-multibyte-characters)
1144 (,buffer (current-buffer)))
1146 (let (default-enable-multibyte-characters)
1147 (set-buffer-multibyte nil)
1149 (set-buffer ,buffer)
1150 (set-buffer-multibyte ,multibyte)))
1151 (let (default-enable-multibyte-characters)
1153 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
1154 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
1156 (defun mm-find-charset-region (b e)
1157 "Return a list of Emacs charsets in the region B to E."
1159 ((and (mm-multibyte-p)
1160 (fboundp 'find-charset-region))
1161 ;; Remove composition since the base charsets have been included.
1162 ;; Remove eight-bit-*, treat them as ascii.
1163 (let ((css (find-charset-region b e)))
1165 '(composition eight-bit-control eight-bit-graphic control-1)
1167 (setq css (delq cs css)))))
1169 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
1172 (narrow-to-region b e)
1173 (goto-char (point-min))
1174 (skip-chars-forward "\0-\177")
1179 (and (boundp 'current-language-environment)
1180 (car (last (assq 'charset
1181 (assoc current-language-environment
1182 language-info-alist))))))
1183 (if (eq charset 'ascii) (setq charset nil))
1186 (car (last (assq mail-parse-charset
1187 mm-mime-mule-charset-alist)))))
1188 (list 'ascii (or charset 'latin-iso8859-1)))))))))
1190 (defun mm-auto-mode-alist ()
1191 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
1192 (let ((alist auto-mode-alist)
1195 (when (listp (cdar alist))
1196 (push (car alist) out))
1200 (defvar mm-inhibit-file-name-handlers
1201 '(jka-compr-handler image-file-handler epa-file-handler)
1202 "A list of handlers doing (un)compression (etc) thingies.")
1204 (defun mm-insert-file-contents (filename &optional visit beg end replace
1206 "Like `insert-file-contents', but only reads in the file.
1207 A buffer may be modified in several ways after reading into the buffer due
1208 to advanced Emacs features, such as file-name-handlers, format decoding,
1209 `find-file-hooks', etc.
1210 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
1211 This function ensures that none of these modifications will take place."
1212 (let* ((format-alist nil)
1213 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
1214 (default-major-mode 'fundamental-mode)
1215 (enable-local-variables nil)
1216 (after-insert-file-functions nil)
1217 (enable-local-eval nil)
1218 (inhibit-file-name-operation (if inhibit
1219 'insert-file-contents
1220 inhibit-file-name-operation))
1221 (inhibit-file-name-handlers
1223 (append mm-inhibit-file-name-handlers
1224 inhibit-file-name-handlers)
1225 inhibit-file-name-handlers))
1226 (ffh (if (boundp 'find-file-hook)
1229 (val (symbol-value ffh)))
1232 (insert-file-contents filename visit beg end replace)
1235 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1236 "Append the contents of the region to the end of file FILENAME.
1237 When called from a function, expects three arguments,
1238 START, END and FILENAME. START and END are buffer positions
1239 saying what text to write.
1240 Optional fourth argument specifies the coding system to use when
1242 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1243 (let ((coding-system-for-write
1244 (or codesys mm-text-coding-system-for-write
1245 mm-text-coding-system))
1246 (inhibit-file-name-operation (if inhibit
1248 inhibit-file-name-operation))
1249 (inhibit-file-name-handlers
1251 (append mm-inhibit-file-name-handlers
1252 inhibit-file-name-handlers)
1253 inhibit-file-name-handlers)))
1254 (write-region start end filename t 'no-message)
1255 (message "Appended to %s" filename)))
1257 (defun mm-write-region (start end filename &optional append visit lockname
1258 coding-system inhibit)
1260 "Like `write-region'.
1261 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1262 (let ((coding-system-for-write
1263 (or coding-system mm-text-coding-system-for-write
1264 mm-text-coding-system))
1265 (inhibit-file-name-operation (if inhibit
1267 inhibit-file-name-operation))
1268 (inhibit-file-name-handlers
1270 (append mm-inhibit-file-name-handlers
1271 inhibit-file-name-handlers)
1272 inhibit-file-name-handlers)))
1273 (write-region start end filename append visit lockname)))
1275 (autoload 'gmm-write-region "gmm-utils")
1277 ;; It is not a MIME function, but some MIME functions use it.
1278 (if (and (fboundp 'make-temp-file)
1280 (let ((def (symbol-function 'make-temp-file)))
1281 (and (byte-code-function-p def)
1282 (setq def (if (fboundp 'compiled-function-arglist)
1284 (eval (list 'compiled-function-arglist def))
1287 (eq (nth 3 def) 'suffix)))))
1288 (defalias 'mm-make-temp-file 'make-temp-file)
1289 ;; Stolen (and modified for XEmacs) from Emacs 22.
1290 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1291 "Create a temporary file.
1292 The returned file name (created by appending some random characters at the end
1293 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1294 is guaranteed to point to a newly created empty file.
1295 You can then use `write-region' to write new data into the file.
1297 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1299 If SUFFIX is non-nil, add that at the end of the file name."
1300 (let ((umask (default-file-modes))
1304 ;; Create temp files with strict access rights. It's easy to
1305 ;; loosen them later, whereas it's impossible to close the
1306 ;; time-window of loose permissions otherwise.
1307 (set-default-file-modes 448)
1308 (while (condition-case err
1314 (if (fboundp 'temp-directory)
1317 temporary-file-directory))))
1319 (setq file (concat file suffix)))
1321 (make-directory file)
1322 ;; NOTE: This is unsafe if Emacs 20
1323 ;; users and XEmacs users don't use
1324 ;; a secure temp directory.
1325 (gmm-write-region "" nil file nil 'silent
1328 (file-already-exists t)
1329 ;; The XEmacs version of `make-directory' issues
1331 (file-error (or (and (featurep 'xemacs)
1332 (file-exists-p file))
1333 (signal (car err) (cdr err)))))
1334 ;; the file was somehow created by someone else between
1335 ;; `make-temp-name' and `write-region', let's try again.
1339 (set-default-file-modes umask)))))
1341 (defun mm-image-load-path (&optional package)
1343 (dolist (path load-path (nreverse result))
1346 (setq dir (concat (file-name-directory
1347 (directory-file-name path))
1348 "etc/images/" (or package "gnus/")))))
1350 (push path result))))
1352 ;; Fixme: This doesn't look useful where it's used.
1353 (if (fboundp 'detect-coding-region)
1354 (defun mm-detect-coding-region (start end)
1355 "Like `detect-coding-region' except returning the best one."
1356 (let ((coding-systems
1357 (detect-coding-region start end)))
1358 (or (car-safe coding-systems)
1360 (defun mm-detect-coding-region (start end)
1361 (let ((point (point)))
1363 (skip-chars-forward "\0-\177" end)
1365 (if (eq (point) end) 'ascii (mm-guess-charset))
1366 (goto-char point)))))
1368 (declare-function mm-detect-coding-region "mm-util" (start end))
1370 (if (fboundp 'coding-system-get)
1371 (defun mm-detect-mime-charset-region (start end)
1372 "Detect MIME charset of the text in the region between START and END."
1373 (let ((cs (mm-detect-coding-region start end)))
1374 (or (coding-system-get cs :mime-charset)
1375 (coding-system-get cs 'mime-charset))))
1376 (defun mm-detect-mime-charset-region (start end)
1377 "Detect MIME charset of the text in the region between START and END."
1378 (let ((cs (mm-detect-coding-region start end)))
1382 (unless (fboundp 'coding-system-to-mime-charset)
1383 (defalias 'coding-system-to-mime-charset 'ignore)))
1385 (defun mm-coding-system-to-mime-charset (coding-system)
1386 "Return the MIME charset corresponding to CODING-SYSTEM.
1387 To make this function work with XEmacs, the APEL package is required."
1389 (or (and (fboundp 'coding-system-get)
1390 (or (coding-system-get coding-system :mime-charset)
1391 (coding-system-get coding-system 'mime-charset)))
1392 (and (featurep 'xemacs)
1393 (or (and (fboundp 'coding-system-to-mime-charset)
1394 (not (eq (symbol-function 'coding-system-to-mime-charset)
1396 (and (condition-case nil
1399 (fboundp 'coding-system-to-mime-charset)))
1400 (coding-system-to-mime-charset coding-system)))))
1403 (require 'jka-compr))
1405 (defun mm-decompress-buffer (filename &optional inplace force)
1406 "Decompress buffer's contents, depending on jka-compr.
1407 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1408 agrees with `jka-compr-compression-info-list', decompression is done.
1409 Signal an error if FORCE is neither nil nor t and compressed data are
1410 not decompressed because `auto-compression-mode' is disabled.
1411 If INPLACE is nil, return decompressed data or nil without modifying
1412 the buffer. Otherwise, replace the buffer's contents with the
1413 decompressed data. The buffer's multibyteness must be turned off."
1416 (prog1 t (require 'jka-compr))
1417 (and (fboundp 'jka-compr-installed-p)
1418 (jka-compr-installed-p))))
1419 (let ((info (jka-compr-get-compression-info filename)))
1421 (unless (or (memq force (list nil t))
1422 (jka-compr-installed-p))
1424 (let ((prog (jka-compr-info-uncompress-program info))
1425 (args (jka-compr-info-uncompress-args info))
1426 (msg (format "%s %s..."
1427 (jka-compr-info-uncompress-message info)
1429 (err-file (jka-compr-make-temp-name))
1430 (cur (current-buffer))
1431 (coding-system-for-read mm-binary-coding-system)
1432 (coding-system-for-write mm-binary-coding-system)
1435 (mm-with-unibyte-buffer
1436 (insert-buffer-substring cur)
1439 (unless (memq (apply 'call-process-region
1440 (point-min) (point-max)
1441 prog t (list t err-file) nil args)
1442 jka-compr-acceptable-retval-list)
1446 (delete "" (split-string
1448 (insert-file-contents err-file)
1454 (format "Error while executing \"%s %s < %s\""
1455 prog (mapconcat 'identity args " ")
1457 (setq retval (buffer-string)))
1459 (setq err-msg (error-message-string err)))))
1460 (when (file-exists-p err-file)
1461 (ignore-errors (jka-compr-delete-temp-file err-file)))
1464 (delete-region (point-min) (point-max))
1467 (message "%s" (or err-msg (concat msg "done")))
1471 (unless (fboundp 'coding-system-name)
1472 (defalias 'coding-system-name 'ignore))
1473 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1474 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1475 (unless (fboundp 'find-operation-coding-system)
1476 (defalias 'find-operation-coding-system 'ignore)))
1478 (defun mm-find-buffer-file-coding-system (&optional filename)
1479 "Find coding system used to decode the contents of the current buffer.
1480 This function looks for the coding system magic cookie or examines the
1481 coding system specified by `file-coding-system-alist' being associated
1482 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1483 gzip, bzip2, etc. are allowed."
1485 (setq filename buffer-file-name))
1487 (let ((decomp (unless ;; No worth to examine charset of tar files.
1490 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1492 (mm-decompress-buffer filename nil t))))
1494 (set-buffer (let (default-enable-multibyte-characters)
1495 (generate-new-buffer " *temp*")))
1497 (setq filename (file-name-sans-extension filename)))
1498 (goto-char (point-min))
1501 ((boundp 'set-auto-coding-function) ;; Emacs
1503 (or (funcall (symbol-value 'set-auto-coding-function)
1504 filename (- (point-max) (point-min)))
1505 (car (find-operation-coding-system 'insert-file-contents
1507 (let (auto-coding-alist)
1509 (funcall (symbol-value 'set-auto-coding-function)
1510 nil (- (point-max) (point-min)))
1512 ((and (featurep 'xemacs) (featurep 'file-coding)) ;; XEmacs
1513 (let ((case-fold-search t)
1514 (end (point-at-eol))
1517 (and (re-search-forward "-\\*-+[\t ]*" end t)
1519 (setq start (match-end 0))
1520 (re-search-forward "[\t ]*-+\\*-" end t))
1522 (setq end (match-beginning 0))
1524 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1526 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1528 (find-coding-system (setq codesys
1529 (intern (match-string 1))))
1531 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1534 (setq start (match-end 0))
1535 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1537 (setq end (match-beginning 0))
1540 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1542 (find-coding-system (setq codesys
1543 (intern (match-string 1))))
1546 (goto-char (point-min))
1547 (setq case-fold-search nil)
1548 (re-search-forward "^;;;coding system: "
1549 ;;(+ (point-min) 3000) t))
1551 (looking-at "[^\t\n\r ]+")
1553 (setq codesys (intern (match-string 0))))
1557 (find-file-coding-system-for-read-from-filename
1559 (coding-system-name (coding-system-base codesys)))))))
1561 (kill-buffer (current-buffer)))))))
1565 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1566 ;;; mm-util.el ends here