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 ;; ISO8859-1 is a bogus name for ISO-8859-1
273 ,@(when (and (not (mm-coding-system-p 'iso8859-1))
274 (mm-coding-system-p 'iso-8859-1))
275 '((iso8859-1 . iso-8859-1)))
277 "A mapping from unknown or invalid charset names to the real charset names.
279 See `mm-codepage-iso-8859-list' and `mm-codepage-ibm-list'.")
281 (defcustom mm-codepage-iso-8859-list
282 (list 1250 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
283 ;; Outlook users in Czech republic. Use this to allow reading of
284 ;; their e-mails. cp1250 should be defined by M-x codepage-setup
286 '(1252 . 1) ;; Windows-1252 is a superset of iso-8859-1 (West
287 ;; Europe). See also `gnus-article-dumbquotes-map'.
288 '(1254 . 9) ;; Windows-1254 is a superset of iso-8859-9 (Turkish).
289 '(1255 . 8));; Windows-1255 is a superset of iso-8859-8 (Hebrew).
290 "A list of Windows codepage numbers and iso-8859 charset numbers.
292 If an element is a number corresponding to a supported windows
293 codepage, appropriate entries to `mm-charset-synonym-alist' are
294 added by `mm-setup-codepage-iso-8859'. An element may also be a
295 cons cell where the car is a codepage number and the cdr is the
296 corresponding number of an iso-8859 charset."
297 :type '(list (set :inline t
298 (const 1250 :tag "Central and East European")
299 (const (1252 . 1) :tag "West European")
300 (const (1254 . 9) :tag "Turkish")
301 (const (1255 . 8) :tag "Hebrew"))
305 (integer :tag "Windows codepage number")
306 (cons (integer :tag "Windows codepage number")
307 (integer :tag "iso-8859 charset number")))))
308 :version "22.1" ;; Gnus 5.10.9
311 (defcustom mm-codepage-ibm-list
312 (list 437 ;; (US etc.)
316 863 ;; (Canadian French)
321 866 ;; (Cyrillic - Russian)
326 ;; In Emacs 23 (unicode), cp... and ibm... are aliases.
327 ;; Cf. http://thread.gmane.org/v9lkng5nwy.fsf@marauder.physik.uni-ulm.de
328 "List of IBM codepage numbers.
330 The codepage mappings slighly differ between IBM and other vendors.
331 See \"ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/IBM/README.TXT\".
333 If an element is a number corresponding to a supported windows
334 codepage, appropriate entries to `mm-charset-synonym-alist' are
335 added by `mm-setup-codepage-ibm'."
336 :type '(list (set :inline t
337 (const 437 :tag "US etc.")
338 (const 860 :tag "Portugal")
339 (const 861 :tag "Iceland")
340 (const 862 :tag "Israel")
341 (const 863 :tag "Canadian French")
342 (const 865 :tag "Nordic")
344 (const 850 :tag "Latin 1")
345 (const 855 :tag "Cyrillic")
346 (const 866 :tag "Cyrillic - Russian")
347 (const 857 :tag "Turkish")
348 (const 864 :tag "Arabic")
349 (const 869 :tag "Greek")
350 (const 874 :tag "Thai"))
353 (integer :tag "Codepage number")))
354 :version "22.1" ;; Gnus 5.10.9
357 (defun mm-setup-codepage-iso-8859 (&optional list)
358 "Add appropriate entries to `mm-charset-synonym-alist'.
359 Unless LIST is given, `mm-codepage-iso-8859-list' is used."
361 (setq list mm-codepage-iso-8859-list))
363 (let (cp windows iso)
365 (setq cp (intern (format "cp%d" (car i)))
366 windows (intern (format "windows-%d" (car i)))
367 iso (intern (format "iso-8859-%d" (cdr i))))
368 (setq cp (intern (format "cp%d" i))
369 windows (intern (format "windows-%d" i))))
370 (unless (mm-coding-system-p windows)
371 (if (mm-coding-system-p cp)
372 (add-to-list 'mm-charset-synonym-alist (cons windows cp))
373 (add-to-list 'mm-charset-synonym-alist (cons windows iso)))))))
375 (defun mm-setup-codepage-ibm (&optional list)
376 "Add appropriate entries to `mm-charset-synonym-alist'.
377 Unless LIST is given, `mm-codepage-ibm-list' is used."
379 (setq list mm-codepage-ibm-list))
380 (dolist (number list)
381 (let ((ibm (intern (format "ibm%d" number)))
382 (cp (intern (format "cp%d" number))))
383 (when (and (not (mm-coding-system-p ibm))
384 (mm-coding-system-p cp))
385 (add-to-list 'mm-charset-synonym-alist (cons ibm cp))))))
388 (mm-setup-codepage-iso-8859)
389 (mm-setup-codepage-ibm)
391 ;; Note: this has to be defined before `mm-charset-to-coding-system'.
392 (defcustom mm-charset-eval-alist
393 (if (featurep 'xemacs)
394 nil ;; I don't know what would be useful for XEmacs.
395 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
396 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
397 (windows-1250 . (mm-codepage-setup 1250 t))
398 (windows-1251 . (mm-codepage-setup 1251 t))
399 (windows-1253 . (mm-codepage-setup 1253 t))
400 (windows-1257 . (mm-codepage-setup 1257 t))))
401 "An alist of (CHARSET . FORM) pairs.
402 If an article is encoded in an unknown CHARSET, FORM is
403 evaluated. This allows to load additional libraries providing
404 charsets on demand. If supported by your Emacs version, you
405 could use `autoload-coding-system' here."
406 :version "22.1" ;; Gnus 5.10.9
407 :type '(list (set :inline t
408 (const (windows-1250 . (mm-codepage-setup 1250 t)))
409 (const (windows-1251 . (mm-codepage-setup 1251 t)))
410 (const (windows-1253 . (mm-codepage-setup 1253 t)))
411 (const (windows-1257 . (mm-codepage-setup 1257 t)))
412 (const (cp850 . (mm-codepage-setup 850 nil))))
415 (cons (symbol :tag "charset")
416 (symbol :tag "form"))))
418 (put 'mm-charset-eval-alist 'risky-local-variable t)
420 ;; Note: this function has to be defined before `mm-charset-override-alist'
421 ;; since it will use this function in order to determine its default value
422 ;; when loading mm-util.elc.
423 (defun mm-charset-to-coding-system (charset &optional lbt
424 allow-override silent)
425 "Return coding-system corresponding to CHARSET.
426 CHARSET is a symbol naming a MIME charset.
427 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
428 used as the line break code type of the coding system.
430 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
431 map undesired charset names to their replacement. This should
432 only be used for decoding, not for encoding.
434 A non-nil value of SILENT means don't issue a warning even if CHARSET
436 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
437 (when (stringp charset)
438 (setq charset (intern (downcase charset))))
440 (setq charset (intern (format "%s-%s" charset lbt))))
444 ;; Running in a non-MULE environment.
445 ((or (null (mm-get-coding-system-list))
446 (not (fboundp 'coding-system-get)))
448 ;; Check override list quite early. Should only used for decoding, not for
451 (let ((cs (cdr (assq charset mm-charset-override-alist))))
452 (and cs (mm-coding-system-p cs) cs))))
454 ((eq charset 'us-ascii)
456 ;; Check to see whether we can handle this charset. (This depends
457 ;; on there being some coding system matching each `mime-charset'
458 ;; property defined, as there should be.)
459 ((and (mm-coding-system-p charset)
460 ;;; Doing this would potentially weed out incorrect charsets.
462 ;;; (eq charset (coding-system-get charset 'mime-charset))
465 ;; Eval expressions from `mm-charset-eval-alist'
466 ((let* ((el (assq charset mm-charset-eval-alist))
473 (condition-case nil (eval form) (error nil))
474 ;; (message "Failed to eval `%s'" form))
475 (mm-coding-system-p cs)
476 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
478 ;; Translate invalid charsets.
479 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
481 (mm-coding-system-p cs)
483 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
486 ;; Last resort: search the coding system list for entries which
487 ;; have the right mime-charset in case the canonical name isn't
488 ;; defined (though it should be).
490 ;; mm-get-coding-system-list returns a list of cs without lbt.
492 (dolist (c (mm-get-coding-system-list))
494 (eq charset (or (coding-system-get c :mime-charset)
495 (coding-system-get c 'mime-charset))))
497 (unless (or silent cs)
498 ;; Warn the user about unknown charset:
499 (if (fboundp 'gnus-message)
500 (gnus-message 7 "Unknown charset: %s" charset)
501 (message "Unknown charset: %s" charset)))
504 ;; Note: `mm-charset-to-coding-system' has to be defined before this.
505 (defcustom mm-charset-override-alist
506 ;; Note: pairs that cannot be used in the Emacs version currently running
509 (iso-8859-1 . windows-1252)
510 (iso-8859-8 . windows-1255)
511 (iso-8859-9 . windows-1254))
512 "A mapping from undesired charset names to their replacement.
514 You may add pairs like (iso-8859-1 . windows-1252) here,
515 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
516 superset of iso-8859-1."
523 (mapcar (lambda (pair)
524 (if (mm-charset-to-coding-system (cdr pair)
528 (iso-8859-1 . windows-1252)
529 (iso-8859-8 . windows-1255)
530 (iso-8859-9 . windows-1254)
531 (undecided . windows-1252)))))
532 (val (copy-sequence (default-value 'mm-charset-override-alist)))
535 (push (if (and (prog1
536 (setq pair (assq (caar val) defaults))
537 (setq defaults (delq pair defaults)))
538 (equal (car val) pair))
541 (const :format "(%v" ,(caar val))
542 (symbol :size 3 :format " . %v)\n" ,(cdar val))))
544 (setq val (cdr val)))
546 (push `(const ,(pop defaults)) rest))
549 `(set :inline t :format "%v" ,@(nreverse rest))
550 `(repeat :inline t :tag "Other options"
552 (symbol :size 3 :format "(%v")
553 (symbol :size 3 :format " . %v)\n")))))))
554 ;; Remove pairs that cannot be used in the Emacs version currently
555 ;; running. Note that this section will be evaluated when loading
557 :set (lambda (symbol value)
560 (mapcar (lambda (pair)
561 (if (mm-charset-to-coding-system (cdr pair)
565 :version "22.1" ;; Gnus 5.10.9
568 (defvar mm-binary-coding-system
570 ((mm-coding-system-p 'binary) 'binary)
571 ((mm-coding-system-p 'no-conversion) 'no-conversion)
573 "100% binary coding system.")
575 (defvar mm-text-coding-system
576 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
577 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
578 (and (mm-coding-system-p 'raw-text) 'raw-text))
579 mm-binary-coding-system)
580 "Text-safe coding system (For removing ^M).")
582 (defvar mm-text-coding-system-for-write nil
583 "Text coding system for write.")
585 (defvar mm-auto-save-coding-system
587 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
588 (if (memq system-type '(windows-nt ms-dos ms-windows))
589 (if (mm-coding-system-p 'utf-8-emacs-dos)
590 'utf-8-emacs-dos mm-binary-coding-system)
592 ((mm-coding-system-p 'emacs-mule)
593 (if (memq system-type '(windows-nt ms-dos ms-windows))
594 (if (mm-coding-system-p 'emacs-mule-dos)
595 'emacs-mule-dos mm-binary-coding-system)
597 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
598 (t mm-binary-coding-system))
599 "Coding system of auto save file.")
601 (defvar mm-universal-coding-system mm-auto-save-coding-system
602 "The universal coding system.")
604 ;; Fixme: some of the cars here aren't valid MIME charsets. That
605 ;; should only matter with XEmacs, though.
606 (defvar mm-mime-mule-charset-alist
608 (iso-8859-1 latin-iso8859-1)
609 (iso-8859-2 latin-iso8859-2)
610 (iso-8859-3 latin-iso8859-3)
611 (iso-8859-4 latin-iso8859-4)
612 (iso-8859-5 cyrillic-iso8859-5)
613 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
614 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
615 ;; charset is koi8-r, not iso-8859-5.
616 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
617 (iso-8859-6 arabic-iso8859-6)
618 (iso-8859-7 greek-iso8859-7)
619 (iso-8859-8 hebrew-iso8859-8)
620 (iso-8859-9 latin-iso8859-9)
621 (iso-8859-14 latin-iso8859-14)
622 (iso-8859-15 latin-iso8859-15)
623 (viscii vietnamese-viscii-lower)
624 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
625 (euc-kr korean-ksc5601)
626 (gb2312 chinese-gb2312)
628 (gb18030 gb18030-2-byte
629 gb18030-4-byte-bmp gb18030-4-byte-smp
630 gb18030-4-byte-ext-1 gb18030-4-byte-ext-2)
631 (big5 chinese-big5-1 chinese-big5-2)
633 (thai-tis620 thai-tis620)
634 (windows-1251 cyrillic-iso8859-5)
635 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
636 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
637 latin-jisx0201 japanese-jisx0208-1978
638 chinese-gb2312 japanese-jisx0208
639 korean-ksc5601 japanese-jisx0212)
640 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
641 latin-jisx0201 japanese-jisx0208-1978
642 chinese-gb2312 japanese-jisx0208
643 korean-ksc5601 japanese-jisx0212
644 chinese-cns11643-1 chinese-cns11643-2)
645 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
646 cyrillic-iso8859-5 greek-iso8859-7
647 latin-jisx0201 japanese-jisx0208-1978
648 chinese-gb2312 japanese-jisx0208
649 korean-ksc5601 japanese-jisx0212
650 chinese-cns11643-1 chinese-cns11643-2
651 chinese-cns11643-3 chinese-cns11643-4
652 chinese-cns11643-5 chinese-cns11643-6
654 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
655 japanese-jisx0213-1 japanese-jisx0213-2)
656 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
657 ,(cond ((fboundp 'unicode-precedence-list)
658 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
659 (unicode-precedence-list)))))
660 ((or (not (fboundp 'charsetp)) ;; non-Mule case
661 (charsetp 'unicode-a)
662 (not (mm-coding-system-p 'mule-utf-8)))
663 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
664 (t ;; If we have utf-8 we're in Mule 5+.
667 (coding-system-get 'mule-utf-8 'safe-charsets))))))
668 "Alist of MIME-charset/MULE-charsets.")
670 (defun mm-enrich-utf-8-by-mule-ucs ()
671 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
672 This function will run when the `un-define' module is loaded under
673 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
674 with Mule charsets. It is completely useless for Emacs."
675 (when (boundp 'unicode-basic-translation-charset-order-list)
681 'unicode-basic-translation-charset-order-list))))
682 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
685 (setq mm-mime-mule-charset-alist
686 (nconc mm-mime-mule-charset-alist
687 (list (cons 'utf-8 val))))))
690 ;; Correct by construction, but should be unnecessary for Emacs:
691 (if (featurep 'xemacs)
692 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
693 (when (and (fboundp 'coding-system-list)
694 (fboundp 'sort-coding-systems))
695 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
699 mime (or (coding-system-get cs :mime-charset); Emacs 23 (unicode)
700 (coding-system-get cs 'mime-charset)))
702 (not (eq t (setq mule
703 (coding-system-get cs 'safe-charsets))))
704 (not (assq mime alist)))
705 (push (cons mime (delq 'ascii mule)) alist)))
706 (setq mm-mime-mule-charset-alist (nreverse alist)))))
708 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
709 "A list of special charsets.
710 Valid elements include:
711 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
712 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
715 (defvar mm-iso-8859-15-compatible
716 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
717 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
718 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
720 (defvar mm-iso-8859-x-to-15-table
721 (and (fboundp 'coding-system-p)
722 (mm-coding-system-p 'iso-8859-15)
725 (if (mm-coding-system-p (car cs))
726 (let ((c (string-to-char
727 (decode-coding-string "\341" (car cs)))))
728 (cons (char-charset c)
731 (decode-coding-string "\341" 'iso-8859-15)) c)
732 (string-to-list (decode-coding-string (car (cdr cs))
735 mm-iso-8859-15-compatible))
736 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
738 (defcustom mm-coding-system-priorities
739 (if (boundp 'current-language-environment)
740 (let ((lang (symbol-value 'current-language-environment)))
741 (cond ((string= lang "Japanese")
742 ;; Japanese users prefer iso-2022-jp to euc-japan or
743 ;; shift_jis, however iso-8859-1 should be used when
744 ;; there are only ASCII text and Latin-1 characters.
745 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
746 "Preferred coding systems for encoding outgoing messages.
748 More than one suitable coding system may be found for some text.
749 By default, the coding system with the highest priority is used
750 to encode outgoing messages (see `sort-coding-systems'). If this
751 variable is set, it overrides the default priority."
753 :type '(repeat (symbol :tag "Coding system"))
757 (defvar mm-use-find-coding-systems-region
758 (fboundp 'find-coding-systems-region)
759 "Use `find-coding-systems-region' to find proper coding systems.
761 Setting it to nil is useful on Emacsen supporting Unicode if sending
762 mail with multiple parts is preferred to sending a Unicode one.")
764 ;;; Internal variables:
768 (defun mm-mule-charset-to-mime-charset (charset)
769 "Return the MIME charset corresponding to the given Mule CHARSET."
770 (if (and (fboundp 'find-coding-systems-for-charsets)
771 (fboundp 'sort-coding-systems))
772 (let ((css (sort (sort-coding-systems
773 (find-coding-systems-for-charsets (list charset)))
774 'mm-sort-coding-systems-predicate))
776 (while (and (not mime)
778 (when (setq cs (pop css))
779 (setq mime (or (coding-system-get cs :mime-charset)
780 (coding-system-get cs 'mime-charset)))))
782 (let ((alist (mapcar (lambda (cs)
783 (assq cs mm-mime-mule-charset-alist))
784 (sort (mapcar 'car mm-mime-mule-charset-alist)
785 'mm-sort-coding-systems-predicate)))
788 (when (memq charset (cdar alist))
789 (setq out (caar alist)
795 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
796 (boundp 'default-enable-multibyte-characters)
797 default-enable-multibyte-characters
798 (fboundp 'set-buffer-multibyte))
799 "True in Emacs with Mule.")
802 (defun mm-enable-multibyte ()
803 "Set the multibyte flag of the current buffer.
804 Only do this if the default value of `enable-multibyte-characters' is
805 non-nil. This is a no-op in XEmacs."
806 (set-buffer-multibyte 'to))
807 (defalias 'mm-enable-multibyte 'ignore))
810 (defun mm-disable-multibyte ()
811 "Unset the multibyte flag of in the current buffer.
812 This is a no-op in XEmacs."
813 (set-buffer-multibyte nil))
814 (defalias 'mm-disable-multibyte 'ignore)))
816 (defun mm-preferred-coding-system (charset)
817 ;; A typo in some Emacs versions.
818 (or (get-charset-property charset 'preferred-coding-system)
819 (get-charset-property charset 'prefered-coding-system)))
821 ;; Mule charsets shouldn't be used.
822 (defsubst mm-guess-charset ()
823 "Guess Mule charset from the language environment."
825 mail-parse-mule-charset ;; cached mule-charset
827 (setq mail-parse-mule-charset
828 (and (boundp 'current-language-environment)
831 (assoc current-language-environment
832 language-info-alist))))))
833 (if (or (not mail-parse-mule-charset)
834 (eq mail-parse-mule-charset 'ascii))
835 (setq mail-parse-mule-charset
836 (or (car (last (assq mail-parse-charset
837 mm-mime-mule-charset-alist)))
840 mail-parse-mule-charset)))
842 (defun mm-charset-after (&optional pos)
843 "Return charset of a character in current buffer at position POS.
844 If POS is nil, it defauls to the current point.
845 If POS is out of range, the value is nil.
846 If the charset is `composition', return the actual one."
847 (let ((char (char-after pos)) charset)
848 (if (< (mm-char-int char) 128)
849 (setq charset 'ascii)
850 ;; charset-after is fake in some Emacsen.
851 (setq charset (and (fboundp 'char-charset) (char-charset char)))
852 (if (eq charset 'composition) ; Mule 4
853 (let ((p (or pos (point))))
854 (cadr (find-charset-region p (1+ p))))
855 (if (and charset (not (memq charset '(ascii eight-bit-control
856 eight-bit-graphic))))
858 (mm-guess-charset))))))
860 (defun mm-mime-charset (charset)
861 "Return the MIME charset corresponding to the given Mule CHARSET."
862 (if (eq charset 'unknown)
863 (error "The message contains non-printable characters, please use attachment"))
864 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
865 ;; This exists in Emacs 20.
867 (and (mm-preferred-coding-system charset)
868 (or (coding-system-get
869 (mm-preferred-coding-system charset) :mime-charset)
871 (mm-preferred-coding-system charset) 'mime-charset)))
872 (and (eq charset 'ascii)
874 (mm-preferred-coding-system charset)
875 (mm-mule-charset-to-mime-charset charset))
876 ;; This is for XEmacs.
877 (mm-mule-charset-to-mime-charset charset)))
879 (if (fboundp 'delete-dups)
880 (defalias 'mm-delete-duplicates 'delete-dups)
881 (defun mm-delete-duplicates (list)
882 "Destructively remove `equal' duplicates from LIST.
883 Store the result in LIST and return it. LIST must be a proper list.
884 Of several `equal' occurrences of an element in LIST, the first
887 This is a compatibility function for Emacsen without `delete-dups'."
888 ;; Code from `subr.el' in Emacs 22:
891 (setcdr tail (delete (car tail) (cdr tail)))
892 (setq tail (cdr tail))))
895 ;; Fixme: This is used in places when it should be testing the
896 ;; default multibyteness. See mm-default-multibyte-p.
898 (if (and (not (featurep 'xemacs))
899 (boundp 'enable-multibyte-characters))
900 (defun mm-multibyte-p ()
901 "Non-nil if multibyte is enabled in the current buffer."
902 enable-multibyte-characters)
903 (defun mm-multibyte-p () (featurep 'mule))))
905 (defun mm-default-multibyte-p ()
906 "Return non-nil if the session is multibyte.
907 This affects whether coding conversion should be attempted generally."
909 (if (boundp 'default-enable-multibyte-characters)
910 default-enable-multibyte-characters
913 (defun mm-iso-8859-x-to-15-region (&optional b e)
914 (if (fboundp 'char-charset)
915 (let (charset item c inconvertible)
917 (if e (narrow-to-region b e))
918 (goto-char (point-min))
919 (skip-chars-forward "\0-\177")
922 ((not (setq item (assq (char-charset (setq c (char-after)))
923 mm-iso-8859-x-to-15-table)))
925 ((memq c (cdr (cdr item)))
926 (setq inconvertible t)
929 (insert-before-markers (prog1 (+ c (car (cdr item)))
931 (skip-chars-forward "\0-\177")))
932 (not inconvertible))))
934 (defun mm-sort-coding-systems-predicate (a b)
937 ;; Note: invalid entries are dropped silently
938 (and (setq cs (mm-coding-system-p cs))
939 (coding-system-base cs)))
940 mm-coding-system-priorities)))
941 (and (setq a (mm-coding-system-p a))
942 (if (setq b (mm-coding-system-p b))
943 (> (length (memq (coding-system-base a) priorities))
944 (length (memq (coding-system-base b) priorities)))
948 (autoload 'latin-unity-massage-name "latin-unity")
949 (autoload 'latin-unity-maybe-remap "latin-unity")
950 (autoload 'latin-unity-representations-feasible-region "latin-unity")
951 (autoload 'latin-unity-representations-present-region "latin-unity"))
953 (defvar latin-unity-coding-systems)
954 (defvar latin-unity-ucs-list)
956 (defun mm-xemacs-find-mime-charset-1 (begin end)
957 "Determine which MIME charset to use to send region as message.
958 This uses the XEmacs-specific latin-unity package to better handle the
959 case where identical characters from diverse ISO-8859-? character sets
960 can be encoded using a single one of the corresponding coding systems.
962 It treats `mm-coding-system-priorities' as the list of preferred
963 coding systems; a useful example setting for this list in Western
964 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
965 to the very standard Latin 1 coding system, and only move to coding
966 systems that are less supported as is necessary to encode the
967 characters that exist in the buffer.
969 Latin Unity doesn't know about those non-ASCII Roman characters that
970 are available in various East Asian character sets. As such, its
971 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
972 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
973 But this is very much a corner case, so don't worry about it."
974 (let ((systems mm-coding-system-priorities) csets psets curset)
976 ;; Load the Latin Unity library, if available.
977 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
978 (require 'latin-unity))
980 ;; Now, can we use it?
981 (if (featurep 'latin-unity)
983 (setq csets (latin-unity-representations-feasible-region begin end)
984 psets (latin-unity-representations-present-region begin end))
988 ;; Pass back the first coding system in the preferred list
989 ;; that can encode the whole region.
990 (dolist (curset systems)
991 (setq curset (latin-unity-massage-name 'buffer-default curset))
993 ;; If the coding system is a universal coding system, then
994 ;; it can certainly encode all the characters in the region.
995 (if (memq curset latin-unity-ucs-list)
996 (throw 'done (list curset)))
998 ;; If a coding system isn't universal, and isn't in
999 ;; the list that latin unity knows about, we can't
1000 ;; decide whether to use it here. Leave that until later
1001 ;; in `mm-find-mime-charset-region' function, whence we
1002 ;; have been called.
1003 (unless (memq curset latin-unity-coding-systems)
1006 ;; Right, we know about this coding system, and it may
1007 ;; conceivably be able to encode all the characters in
1009 (if (latin-unity-maybe-remap begin end curset csets psets t)
1010 (throw 'done (list curset))))
1012 ;; Can't encode using anything from the
1013 ;; `mm-coding-system-priorities' list.
1014 ;; Leave `mm-find-mime-charset' to do most of the work.
1017 ;; Right, latin unity isn't available; let `mm-find-charset-region'
1018 ;; take its default action, which equally applies to GNU Emacs.
1021 (defmacro mm-xemacs-find-mime-charset (begin end)
1022 (when (featurep 'xemacs)
1023 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
1025 (declare-function mm-delete-duplicates "mm-util" (list))
1027 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
1028 "Return the MIME charsets needed to encode the region between B and E.
1029 nil means ASCII, a single-element list represents an appropriate MIME
1030 charset, and a longer list means no appropriate charset."
1032 ;; The return possibilities of this function are a mess...
1033 (or (and (mm-multibyte-p)
1034 mm-use-find-coding-systems-region
1035 ;; Find the mime-charset of the most preferred coding
1036 ;; system that has one.
1037 (let ((systems (find-coding-systems-region b e)))
1038 (when mm-coding-system-priorities
1040 (sort systems 'mm-sort-coding-systems-predicate)))
1041 (setq systems (delq 'compound-text systems))
1042 (unless (equal systems '(undecided))
1044 (let* ((head (pop systems))
1045 (cs (or (coding-system-get head :mime-charset)
1046 (coding-system-get head 'mime-charset))))
1047 ;; The mime-charset (`x-ctext') of
1048 ;; `compound-text' is not in the IANA list. We
1049 ;; shouldn't normally use anything here with a
1050 ;; mime-charset having an `x-' prefix.
1051 ;; Fixme: Allow this to be overridden, since
1052 ;; there is existing use of x-ctext.
1053 ;; Also people apparently need the coding system
1054 ;; `iso-2022-jp-3' (which Mule-UCS defines with
1055 ;; mime-charset, though it's not valid).
1057 (not (string-match "^[Xx]-" (symbol-name cs)))
1058 ;; UTF-16 of any variety is invalid for
1059 ;; text parts and, unfortunately, has
1060 ;; mime-charset defined both in Mule-UCS
1061 ;; and versions of Emacs. (The name
1062 ;; might be `mule-utf-16...' or
1064 (not (string-match "utf-16" (symbol-name cs))))
1066 charsets (list cs))))))
1068 ;; If we're XEmacs, and some coding system is appropriate,
1069 ;; mm-xemacs-find-mime-charset will return an appropriate list.
1070 ;; Otherwise, we'll get nil, and the next setq will get invoked.
1071 (setq charsets (mm-xemacs-find-mime-charset b e))
1073 ;; Fixme: won't work for unibyte Emacs 23:
1075 ;; We're not multibyte, or a single coding system won't cover it.
1077 (mm-delete-duplicates
1078 (mapcar 'mm-mime-charset
1080 (mm-find-charset-region b e))))))
1081 (if (and (> (length charsets) 1)
1082 (memq 'iso-8859-15 charsets)
1083 (memq 'iso-8859-15 hack-charsets)
1084 (save-excursion (mm-iso-8859-x-to-15-region b e)))
1085 (dolist (x mm-iso-8859-15-compatible)
1086 (setq charsets (delq (car x) charsets))))
1087 (if (and (memq 'iso-2022-jp-2 charsets)
1088 (memq 'iso-2022-jp-2 hack-charsets))
1089 (setq charsets (delq 'iso-2022-jp charsets)))
1090 ;; Attempt to reduce the number of charsets if utf-8 is available.
1091 (if (and (featurep 'xemacs)
1092 (> (length charsets) 1)
1093 (mm-coding-system-p 'utf-8))
1094 (let ((mm-coding-system-priorities
1095 (cons 'utf-8 mm-coding-system-priorities)))
1097 (mm-delete-duplicates
1098 (mapcar 'mm-mime-charset
1100 (mm-find-charset-region b e)))))))
1103 (defmacro mm-with-unibyte-buffer (&rest forms)
1104 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1105 Use unibyte mode for this."
1107 (mm-disable-multibyte)
1109 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
1110 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
1112 (defmacro mm-with-multibyte-buffer (&rest forms)
1113 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1114 Use multibyte mode for this."
1116 (mm-enable-multibyte)
1118 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
1119 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
1121 (defmacro mm-with-unibyte-current-buffer (&rest forms)
1122 "Evaluate FORMS with current buffer temporarily made unibyte.
1123 Also bind `default-enable-multibyte-characters' to nil.
1124 Equivalent to `progn' in XEmacs
1126 NOTE: Use this macro with caution in multibyte buffers (it is not
1127 worth using this macro in unibyte buffers of course). Use of
1128 `(set-buffer-multibyte t)', which is run finally, is generally
1129 harmful since it is likely to modify existing data in the buffer.
1130 For instance, it converts \"\\300\\255\" into \"\\255\" in
1131 Emacs 23 (unicode)."
1132 (let ((multibyte (make-symbol "multibyte"))
1133 (buffer (make-symbol "buffer")))
1135 (let ((,multibyte enable-multibyte-characters)
1136 (,buffer (current-buffer)))
1138 (let (default-enable-multibyte-characters)
1139 (set-buffer-multibyte nil)
1141 (set-buffer ,buffer)
1142 (set-buffer-multibyte ,multibyte)))
1143 (let (default-enable-multibyte-characters)
1145 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
1146 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
1148 (defun mm-find-charset-region (b e)
1149 "Return a list of Emacs charsets in the region B to E."
1151 ((and (mm-multibyte-p)
1152 (fboundp 'find-charset-region))
1153 ;; Remove composition since the base charsets have been included.
1154 ;; Remove eight-bit-*, treat them as ascii.
1155 (let ((css (find-charset-region b e)))
1157 '(composition eight-bit-control eight-bit-graphic control-1)
1159 (setq css (delq cs css)))))
1161 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
1164 (narrow-to-region b e)
1165 (goto-char (point-min))
1166 (skip-chars-forward "\0-\177")
1171 (and (boundp 'current-language-environment)
1172 (car (last (assq 'charset
1173 (assoc current-language-environment
1174 language-info-alist))))))
1175 (if (eq charset 'ascii) (setq charset nil))
1178 (car (last (assq mail-parse-charset
1179 mm-mime-mule-charset-alist)))))
1180 (list 'ascii (or charset 'latin-iso8859-1)))))))))
1182 (defun mm-auto-mode-alist ()
1183 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
1184 (let ((alist auto-mode-alist)
1187 (when (listp (cdar alist))
1188 (push (car alist) out))
1192 (defvar mm-inhibit-file-name-handlers
1193 '(jka-compr-handler image-file-handler epa-file-handler)
1194 "A list of handlers doing (un)compression (etc) thingies.")
1196 (defun mm-insert-file-contents (filename &optional visit beg end replace
1198 "Like `insert-file-contents', but only reads in the file.
1199 A buffer may be modified in several ways after reading into the buffer due
1200 to advanced Emacs features, such as file-name-handlers, format decoding,
1201 `find-file-hooks', etc.
1202 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
1203 This function ensures that none of these modifications will take place."
1204 (let* ((format-alist nil)
1205 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
1206 (default-major-mode 'fundamental-mode)
1207 (enable-local-variables nil)
1208 (after-insert-file-functions nil)
1209 (enable-local-eval nil)
1210 (inhibit-file-name-operation (if inhibit
1211 'insert-file-contents
1212 inhibit-file-name-operation))
1213 (inhibit-file-name-handlers
1215 (append mm-inhibit-file-name-handlers
1216 inhibit-file-name-handlers)
1217 inhibit-file-name-handlers))
1218 (ffh (if (boundp 'find-file-hook)
1221 (val (symbol-value ffh)))
1224 (insert-file-contents filename visit beg end replace)
1227 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1228 "Append the contents of the region to the end of file FILENAME.
1229 When called from a function, expects three arguments,
1230 START, END and FILENAME. START and END are buffer positions
1231 saying what text to write.
1232 Optional fourth argument specifies the coding system to use when
1234 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1235 (let ((coding-system-for-write
1236 (or codesys mm-text-coding-system-for-write
1237 mm-text-coding-system))
1238 (inhibit-file-name-operation (if inhibit
1240 inhibit-file-name-operation))
1241 (inhibit-file-name-handlers
1243 (append mm-inhibit-file-name-handlers
1244 inhibit-file-name-handlers)
1245 inhibit-file-name-handlers)))
1246 (write-region start end filename t 'no-message)
1247 (message "Appended to %s" filename)))
1249 (defun mm-write-region (start end filename &optional append visit lockname
1250 coding-system inhibit)
1252 "Like `write-region'.
1253 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1254 (let ((coding-system-for-write
1255 (or coding-system mm-text-coding-system-for-write
1256 mm-text-coding-system))
1257 (inhibit-file-name-operation (if inhibit
1259 inhibit-file-name-operation))
1260 (inhibit-file-name-handlers
1262 (append mm-inhibit-file-name-handlers
1263 inhibit-file-name-handlers)
1264 inhibit-file-name-handlers)))
1265 (write-region start end filename append visit lockname)))
1267 (autoload 'gmm-write-region "gmm-utils")
1269 ;; It is not a MIME function, but some MIME functions use it.
1270 (if (and (fboundp 'make-temp-file)
1272 (let ((def (symbol-function 'make-temp-file)))
1273 (and (byte-code-function-p def)
1274 (setq def (if (fboundp 'compiled-function-arglist)
1276 (eval (list 'compiled-function-arglist def))
1279 (eq (nth 3 def) 'suffix)))))
1280 (defalias 'mm-make-temp-file 'make-temp-file)
1281 ;; Stolen (and modified for XEmacs) from Emacs 22.
1282 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1283 "Create a temporary file.
1284 The returned file name (created by appending some random characters at the end
1285 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1286 is guaranteed to point to a newly created empty file.
1287 You can then use `write-region' to write new data into the file.
1289 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1291 If SUFFIX is non-nil, add that at the end of the file name."
1292 (let ((umask (default-file-modes))
1296 ;; Create temp files with strict access rights. It's easy to
1297 ;; loosen them later, whereas it's impossible to close the
1298 ;; time-window of loose permissions otherwise.
1299 (set-default-file-modes 448)
1300 (while (condition-case err
1306 (if (fboundp 'temp-directory)
1309 temporary-file-directory))))
1311 (setq file (concat file suffix)))
1313 (make-directory file)
1314 ;; NOTE: This is unsafe if Emacs 20
1315 ;; users and XEmacs users don't use
1316 ;; a secure temp directory.
1317 (gmm-write-region "" nil file nil 'silent
1320 (file-already-exists t)
1321 ;; The XEmacs version of `make-directory' issues
1323 (file-error (or (and (featurep 'xemacs)
1324 (file-exists-p file))
1325 (signal (car err) (cdr err)))))
1326 ;; the file was somehow created by someone else between
1327 ;; `make-temp-name' and `write-region', let's try again.
1331 (set-default-file-modes umask)))))
1333 (defun mm-image-load-path (&optional package)
1335 (dolist (path load-path (nreverse result))
1338 (setq dir (concat (file-name-directory
1339 (directory-file-name path))
1340 "etc/images/" (or package "gnus/")))))
1342 (push path result))))
1344 ;; Fixme: This doesn't look useful where it's used.
1345 (if (fboundp 'detect-coding-region)
1346 (defun mm-detect-coding-region (start end)
1347 "Like `detect-coding-region' except returning the best one."
1348 (let ((coding-systems
1349 (detect-coding-region start end)))
1350 (or (car-safe coding-systems)
1352 (defun mm-detect-coding-region (start end)
1353 (let ((point (point)))
1355 (skip-chars-forward "\0-\177" end)
1357 (if (eq (point) end) 'ascii (mm-guess-charset))
1358 (goto-char point)))))
1360 (declare-function mm-detect-coding-region "mm-util" (start end))
1362 (if (fboundp 'coding-system-get)
1363 (defun mm-detect-mime-charset-region (start end)
1364 "Detect MIME charset of the text in the region between START and END."
1365 (let ((cs (mm-detect-coding-region start end)))
1366 (or (coding-system-get cs :mime-charset)
1367 (coding-system-get cs 'mime-charset))))
1368 (defun mm-detect-mime-charset-region (start end)
1369 "Detect MIME charset of the text in the region between START and END."
1370 (let ((cs (mm-detect-coding-region start end)))
1374 (unless (fboundp 'coding-system-to-mime-charset)
1375 (defalias 'coding-system-to-mime-charset 'ignore)))
1377 (defun mm-coding-system-to-mime-charset (coding-system)
1378 "Return the MIME charset corresponding to CODING-SYSTEM.
1379 To make this function work with XEmacs, the APEL package is required."
1381 (or (and (fboundp 'coding-system-get)
1382 (or (coding-system-get coding-system :mime-charset)
1383 (coding-system-get coding-system 'mime-charset)))
1384 (and (featurep 'xemacs)
1385 (or (and (fboundp 'coding-system-to-mime-charset)
1386 (not (eq (symbol-function 'coding-system-to-mime-charset)
1388 (and (condition-case nil
1391 (fboundp 'coding-system-to-mime-charset)))
1392 (coding-system-to-mime-charset coding-system)))))
1395 (require 'jka-compr))
1397 (defun mm-decompress-buffer (filename &optional inplace force)
1398 "Decompress buffer's contents, depending on jka-compr.
1399 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1400 agrees with `jka-compr-compression-info-list', decompression is done.
1401 Signal an error if FORCE is neither nil nor t and compressed data are
1402 not decompressed because `auto-compression-mode' is disabled.
1403 If INPLACE is nil, return decompressed data or nil without modifying
1404 the buffer. Otherwise, replace the buffer's contents with the
1405 decompressed data. The buffer's multibyteness must be turned off."
1408 (prog1 t (require 'jka-compr))
1409 (and (fboundp 'jka-compr-installed-p)
1410 (jka-compr-installed-p))))
1411 (let ((info (jka-compr-get-compression-info filename)))
1413 (unless (or (memq force (list nil t))
1414 (jka-compr-installed-p))
1416 (let ((prog (jka-compr-info-uncompress-program info))
1417 (args (jka-compr-info-uncompress-args info))
1418 (msg (format "%s %s..."
1419 (jka-compr-info-uncompress-message info)
1421 (err-file (jka-compr-make-temp-name))
1422 (cur (current-buffer))
1423 (coding-system-for-read mm-binary-coding-system)
1424 (coding-system-for-write mm-binary-coding-system)
1427 (mm-with-unibyte-buffer
1428 (insert-buffer-substring cur)
1431 (unless (memq (apply 'call-process-region
1432 (point-min) (point-max)
1433 prog t (list t err-file) nil args)
1434 jka-compr-acceptable-retval-list)
1438 (delete "" (split-string
1440 (insert-file-contents err-file)
1446 (format "Error while executing \"%s %s < %s\""
1447 prog (mapconcat 'identity args " ")
1449 (setq retval (buffer-string)))
1451 (setq err-msg (error-message-string err)))))
1452 (when (file-exists-p err-file)
1453 (ignore-errors (jka-compr-delete-temp-file err-file)))
1456 (delete-region (point-min) (point-max))
1459 (message "%s" (or err-msg (concat msg "done")))
1463 (unless (fboundp 'coding-system-name)
1464 (defalias 'coding-system-name 'ignore))
1465 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1466 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1467 (unless (fboundp 'find-operation-coding-system)
1468 (defalias 'find-operation-coding-system 'ignore)))
1470 (defun mm-find-buffer-file-coding-system (&optional filename)
1471 "Find coding system used to decode the contents of the current buffer.
1472 This function looks for the coding system magic cookie or examines the
1473 coding system specified by `file-coding-system-alist' being associated
1474 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1475 gzip, bzip2, etc. are allowed."
1477 (setq filename buffer-file-name))
1479 (let ((decomp (unless ;; No worth to examine charset of tar files.
1482 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1484 (mm-decompress-buffer filename nil t))))
1486 (set-buffer (let (default-enable-multibyte-characters)
1487 (generate-new-buffer " *temp*")))
1489 (setq filename (file-name-sans-extension filename)))
1490 (goto-char (point-min))
1493 ((boundp 'set-auto-coding-function) ;; Emacs
1495 (or (funcall (symbol-value 'set-auto-coding-function)
1496 filename (- (point-max) (point-min)))
1497 (car (find-operation-coding-system 'insert-file-contents
1499 (let (auto-coding-alist)
1501 (funcall (symbol-value 'set-auto-coding-function)
1502 nil (- (point-max) (point-min)))
1504 ((and (featurep 'xemacs) (featurep 'file-coding)) ;; XEmacs
1505 (let ((case-fold-search t)
1506 (end (point-at-eol))
1509 (and (re-search-forward "-\\*-+[\t ]*" end t)
1511 (setq start (match-end 0))
1512 (re-search-forward "[\t ]*-+\\*-" end t))
1514 (setq end (match-beginning 0))
1516 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1518 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1520 (find-coding-system (setq codesys
1521 (intern (match-string 1))))
1523 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1526 (setq start (match-end 0))
1527 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1529 (setq end (match-beginning 0))
1532 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1534 (find-coding-system (setq codesys
1535 (intern (match-string 1))))
1538 (goto-char (point-min))
1539 (setq case-fold-search nil)
1540 (re-search-forward "^;;;coding system: "
1541 ;;(+ (point-min) 3000) t))
1543 (looking-at "[^\t\n\r ]+")
1545 (setq codesys (intern (match-string 0))))
1549 (find-file-coding-system-for-read-from-filename
1551 (coding-system-name (coding-system-base codesys)))))))
1553 (kill-buffer (current-buffer)))))))
1557 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1558 ;;; mm-util.el ends here