1 ;;; mm-util.el --- Utility functions for Mule and low level things
3 ;; Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004,
4 ;; 2005 Free Software Foundation, Inc.
6 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
7 ;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
8 ;; This file is part of GNU Emacs.
10 ;; GNU Emacs is free software; you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation; either version 2, or (at your option)
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs; see the file COPYING. If not, write to the
22 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
23 ;; Boston, MA 02110-1301, USA.
29 (eval-when-compile (require 'cl))
35 (let ((nfunc (intern (format "mm-%s" (car elem)))))
36 (if (fboundp (car elem))
37 (defalias nfunc (car elem))
38 (defalias nfunc (cdr elem)))))
39 '((decode-coding-string . (lambda (s a) s))
40 (encode-coding-string . (lambda (s a) s))
41 (encode-coding-region . ignore)
42 (coding-system-list . ignore)
43 (decode-coding-region . ignore)
45 (coding-system-equal . equal)
46 (annotationp . ignore)
47 (set-buffer-file-coding-system . ignore)
54 (mapcar (lambda (e) (list (symbol-name (car e))))
55 mm-mime-mule-charset-alist)
58 . (lambda (from to string &optional inplace)
59 ;; stolen (and renamed) from nnheader.el
60 "Replace characters in STRING from FROM to TO.
61 Unless optional argument INPLACE is non-nil, return a new string."
62 (let ((string (if inplace string (copy-sequence string)))
65 ;; Replace all occurrences of FROM with TO.
67 (when (= (aref string idx) from)
72 . (lambda (string regexp rep &optional literal)
73 "See `replace-regexp-in-string', only the order of args differs."
74 (replace-regexp-in-string regexp rep string nil literal)))
75 (string-as-unibyte . identity)
76 (string-make-unibyte . identity)
77 ;; string-as-multibyte often doesn't really do what you think it does.
79 ;; (aref (string-as-multibyte "\201") 0) -> 129 (aka ?\201)
80 ;; (aref (string-as-multibyte "\300") 0) -> 192 (aka ?\300)
81 ;; (aref (string-as-multibyte "\300\201") 0) -> 192 (aka ?\300)
82 ;; (aref (string-as-multibyte "\300\201") 1) -> 129 (aka ?\201)
84 ;; (aref (string-as-multibyte "\201\300") 0) -> 2240
85 ;; (aref (string-as-multibyte "\201\300") 1) -> <error>
86 ;; Better use string-to-multibyte or encode-coding-string.
87 ;; If you really need string-as-multibyte somewhere it's usually
88 ;; because you're using the internal emacs-mule representation (maybe
89 ;; because you're using string-as-unibyte somewhere), which is
90 ;; generally a problem in itself.
91 ;; Here is an approximate equivalence table to help think about it:
92 ;; (string-as-multibyte s) ~= (decode-coding-string s 'emacs-mule)
93 ;; (string-to-multibyte s) ~= (decode-coding-string s 'binary)
94 ;; (string-make-multibyte s) ~= (decode-coding-string s locale-coding-system)
95 (string-as-multibyte . identity)
98 "Return a multibyte string with the same individual chars as string."
100 (lambda (ch) (mm-string-as-multibyte (char-to-string ch)))
102 (multibyte-string-p . ignore)
103 ;; It is not a MIME function, but some MIME functions use it.
104 (make-temp-file . (lambda (prefix &optional dir-flag)
105 (let ((file (expand-file-name
106 (make-temp-name prefix)
107 (if (fboundp 'temp-directory)
109 temporary-file-directory))))
111 (make-directory file))
113 (insert-byte . insert-char)
114 (multibyte-char-to-unibyte . identity))))
117 (defalias 'mm-char-or-char-int-p
119 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
120 ((fboundp 'char-valid-p) 'char-valid-p)
123 ;; Fixme: This seems always to be used to read a MIME charset, so it
124 ;; should be re-named and fixed (in Emacs) to offer completion only on
125 ;; proper charset names (base coding systems which have a
126 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
128 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
129 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
130 ;; Actually, there should be an `mm-coding-system-mime-charset'.
132 (defalias 'mm-read-coding-system
134 ((fboundp 'read-coding-system)
135 (if (and (featurep 'xemacs)
136 (<= (string-to-number emacs-version) 21.1))
137 (lambda (prompt &optional default-coding-system)
138 (read-coding-system prompt))
139 'read-coding-system))
140 (t (lambda (prompt &optional default-coding-system)
141 "Prompt the user for a coding system."
143 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
144 mm-mime-mule-charset-alist)))))))
146 (defvar mm-coding-system-list nil)
147 (defun mm-get-coding-system-list ()
148 "Get the coding system list."
149 (or mm-coding-system-list
150 (setq mm-coding-system-list (mm-coding-system-list))))
152 (defun mm-coding-system-p (cs)
153 "Return non-nil if CS is a symbol naming a coding system.
154 In XEmacs, also return non-nil if CS is a coding system object.
155 If CS is available, return CS itself in Emacs, and return a coding
156 system object in XEmacs."
157 (if (fboundp 'find-coding-system)
158 (and cs (find-coding-system cs))
159 (if (fboundp 'coding-system-p)
160 (when (coding-system-p cs)
162 ;; Is this branch ever actually useful?
163 (car (memq cs (mm-get-coding-system-list))))))
165 (defun mm-codepage-setup (number &optional alias)
166 "Create a coding system cpNUMBER.
167 The coding system is created using `codepage-setup'. If ALIAS is
168 non-nil, an alias is created and added to
169 `mm-charset-synonym-alist'. If ALIAS is a string, it's used as
170 the alias. Else windows-NUMBER is used."
172 (let ((completion-ignore-case t)
173 (candidates (cp-supported-codepages)))
174 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
175 nil t nil nil "437"))))
177 (setq alias (if (stringp alias)
179 (intern (format "windows-%s" number)))))
180 (let* ((cp (intern (format "cp%s" number))))
181 (unless (mm-coding-system-p cp)
182 (codepage-setup number))
184 ;; Don't add alias if setup of cp failed.
185 (mm-coding-system-p cp))
186 (add-to-list 'mm-charset-synonym-alist (cons alias cp)))))
188 (defvar mm-charset-synonym-alist
190 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
191 ,@(unless (mm-coding-system-p 'x-ctext)
192 '((x-ctext . ctext)))
193 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_!
194 ,@(unless (mm-coding-system-p 'iso-8859-15)
195 '((iso-8859-15 . iso-8859-1)))
196 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
197 ,@(unless (mm-coding-system-p 'big5-hkscs)
198 '((big5-hkscs . big5)))
199 ;; Windows-1252 is actually a superset of Latin-1. See also
200 ;; `gnus-article-dumbquotes-map'.
201 ,@(unless (mm-coding-system-p 'windows-1252)
202 (if (mm-coding-system-p 'cp1252)
203 '((windows-1252 . cp1252))
204 '((windows-1252 . iso-8859-1))))
205 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
206 ;; Outlook users in Czech republic. Use this to allow reading of their
207 ;; e-mails. cp1250 should be defined by M-x codepage-setup.
208 ,@(if (and (not (mm-coding-system-p 'windows-1250))
209 (mm-coding-system-p 'cp1250))
210 '((windows-1250 . cp1250)))
211 ;; A Microsoft misunderstanding.
212 ,@(if (and (not (mm-coding-system-p 'unicode))
213 (mm-coding-system-p 'utf-16-le))
214 '((unicode . utf-16-le)))
215 ;; A Microsoft misunderstanding.
216 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
217 (if (mm-coding-system-p 'cp949)
218 '((ks_c_5601-1987 . cp949))
219 '((ks_c_5601-1987 . euc-kr))))
221 "A mapping from unknown or invalid charset names to the real charset names.")
223 (defcustom mm-charset-override-alist
224 `((iso-8859-1 . windows-1252))
225 "A mapping from undesired charset names to their replacement.
227 You may add pair like (iso-8859-1 . windows-1252) here,
228 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
229 superset of iso-8859-1."
230 :type '(list (set :inline t
231 (const (iso-8859-1 . windows-1252))
232 (const (undecided . windows-1252)))
235 (cons (symbol :tag "From charset")
236 (symbol :tag "To charset"))))
237 :version "23.0" ;; No Gnus
240 (defcustom mm-charset-eval-alist
241 (if (featurep 'xemacs)
242 nil ;; I don't know what would be useful for XEmacs.
243 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
244 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
245 (windows-1250 . (mm-codepage-setup 1250 t))
246 (windows-1251 . (mm-codepage-setup 1251 t))
247 (windows-1253 . (mm-codepage-setup 1253 t))
248 (windows-1257 . (mm-codepage-setup 1257 t))))
249 "An alist of (CHARSET . FORM) pairs.
250 If an article is encoded in an unknown CHARSET, FORM is
251 evaluated. This allows to load additional libraries providing
252 charsets on demand. If supported by your Emacs version, you
253 could use `autoload-coding-system' here."
254 :version "23.0" ;; No Gnus
255 :type '(list (set :inline t
256 (const (windows-1250 . (mm-codepage-setup 1250 t)))
257 (const (windows-1251 . (mm-codepage-setup 1251 t)))
258 (const (windows-1253 . (mm-codepage-setup 1253 t)))
259 (const (windows-1257 . (mm-codepage-setup 1257 t)))
260 (const (cp850 . (mm-codepage-setup 850 nil))))
263 (cons (symbol :tag "charset")
264 (symbol :tag "form"))))
267 (defvar mm-binary-coding-system
269 ((mm-coding-system-p 'binary) 'binary)
270 ((mm-coding-system-p 'no-conversion) 'no-conversion)
272 "100% binary coding system.")
274 (defvar mm-text-coding-system
275 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
276 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
277 (and (mm-coding-system-p 'raw-text) 'raw-text))
278 mm-binary-coding-system)
279 "Text-safe coding system (For removing ^M).")
281 (defvar mm-text-coding-system-for-write nil
282 "Text coding system for write.")
284 (defvar mm-auto-save-coding-system
286 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
287 (if (memq system-type '(windows-nt ms-dos ms-windows))
288 (if (mm-coding-system-p 'utf-8-emacs-dos)
289 'utf-8-emacs-dos mm-binary-coding-system)
291 ((mm-coding-system-p 'emacs-mule)
292 (if (memq system-type '(windows-nt ms-dos ms-windows))
293 (if (mm-coding-system-p 'emacs-mule-dos)
294 'emacs-mule-dos mm-binary-coding-system)
296 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
297 (t mm-binary-coding-system))
298 "Coding system of auto save file.")
300 (defvar mm-universal-coding-system mm-auto-save-coding-system
301 "The universal coding system.")
303 ;; Fixme: some of the cars here aren't valid MIME charsets. That
304 ;; should only matter with XEmacs, though.
305 (defvar mm-mime-mule-charset-alist
307 (iso-8859-1 latin-iso8859-1)
308 (iso-8859-2 latin-iso8859-2)
309 (iso-8859-3 latin-iso8859-3)
310 (iso-8859-4 latin-iso8859-4)
311 (iso-8859-5 cyrillic-iso8859-5)
312 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
313 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
314 ;; charset is koi8-r, not iso-8859-5.
315 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
316 (iso-8859-6 arabic-iso8859-6)
317 (iso-8859-7 greek-iso8859-7)
318 (iso-8859-8 hebrew-iso8859-8)
319 (iso-8859-9 latin-iso8859-9)
320 (iso-8859-14 latin-iso8859-14)
321 (iso-8859-15 latin-iso8859-15)
322 (viscii vietnamese-viscii-lower)
323 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
324 (euc-kr korean-ksc5601)
325 (gb2312 chinese-gb2312)
326 (big5 chinese-big5-1 chinese-big5-2)
328 (thai-tis620 thai-tis620)
329 (windows-1251 cyrillic-iso8859-5)
330 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
331 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
332 latin-jisx0201 japanese-jisx0208-1978
333 chinese-gb2312 japanese-jisx0208
334 korean-ksc5601 japanese-jisx0212)
335 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
336 latin-jisx0201 japanese-jisx0208-1978
337 chinese-gb2312 japanese-jisx0208
338 korean-ksc5601 japanese-jisx0212
339 chinese-cns11643-1 chinese-cns11643-2)
340 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
341 cyrillic-iso8859-5 greek-iso8859-7
342 latin-jisx0201 japanese-jisx0208-1978
343 chinese-gb2312 japanese-jisx0208
344 korean-ksc5601 japanese-jisx0212
345 chinese-cns11643-1 chinese-cns11643-2
346 chinese-cns11643-3 chinese-cns11643-4
347 chinese-cns11643-5 chinese-cns11643-6
349 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
350 japanese-jisx0213-1 japanese-jisx0213-2)
351 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
352 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
353 (charsetp 'unicode-a)
354 (not (mm-coding-system-p 'mule-utf-8)))
355 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
356 ;; If we have utf-8 we're in Mule 5+.
359 (coding-system-get 'mule-utf-8 'safe-charsets)))))
360 "Alist of MIME-charset/MULE-charsets.")
362 (defun mm-enrich-utf-8-by-mule-ucs ()
363 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
364 This function will run when the `un-define' module is loaded under
365 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
366 with Mule charsets. It is completely useless for Emacs."
367 (unless (cdr (delete '(mm-enrich-utf-8-by-mule-ucs)
368 (assoc "un-define" after-load-alist)))
369 (setq after-load-alist
370 (delete '("un-define") after-load-alist)))
371 (when (boundp 'unicode-basic-translation-charset-order-list)
377 'unicode-basic-translation-charset-order-list))))
378 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
381 (setq mm-mime-mule-charset-alist
382 (nconc mm-mime-mule-charset-alist
383 (list (cons 'utf-8 val))))))
386 ;; Correct by construction, but should be unnecessary for Emacs:
387 (if (featurep 'xemacs)
388 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
389 (when (and (fboundp 'coding-system-list)
390 (fboundp 'sort-coding-systems))
391 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
395 mime (or (coding-system-get cs :mime-charset) ; Emacs 22
396 (coding-system-get cs 'mime-charset)))
398 (not (eq t (setq mule
399 (coding-system-get cs 'safe-charsets))))
400 (not (assq mime alist)))
401 (push (cons mime (delq 'ascii mule)) alist)))
402 (setq mm-mime-mule-charset-alist (nreverse alist)))))
404 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
405 "A list of special charsets.
406 Valid elements include:
407 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
408 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
411 (defvar mm-iso-8859-15-compatible
412 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
413 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
414 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
416 (defvar mm-iso-8859-x-to-15-table
417 (and (fboundp 'coding-system-p)
418 (mm-coding-system-p 'iso-8859-15)
421 (if (mm-coding-system-p (car cs))
422 (let ((c (string-to-char
423 (decode-coding-string "\341" (car cs)))))
424 (cons (char-charset c)
427 (decode-coding-string "\341" 'iso-8859-15)) c)
428 (string-to-list (decode-coding-string (car (cdr cs))
431 mm-iso-8859-15-compatible))
432 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
434 (defcustom mm-coding-system-priorities
435 (if (boundp 'current-language-environment)
436 (let ((lang (symbol-value 'current-language-environment)))
437 (cond ((string= lang "Japanese")
438 ;; Japanese users prefer iso-2022-jp to euc-japan or
439 ;; shift_jis, however iso-8859-1 should be used when
440 ;; there are only ASCII text and Latin-1 characters.
441 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
442 "Preferred coding systems for encoding outgoing messages.
444 More than one suitable coding system may be found for some text.
445 By default, the coding system with the highest priority is used
446 to encode outgoing messages (see `sort-coding-systems'). If this
447 variable is set, it overrides the default priority."
449 :type '(repeat (symbol :tag "Coding system"))
453 (defvar mm-use-find-coding-systems-region
454 (fboundp 'find-coding-systems-region)
455 "Use `find-coding-systems-region' to find proper coding systems.
457 Setting it to nil is useful on Emacsen supporting Unicode if sending
458 mail with multiple parts is preferred to sending a Unicode one.")
460 ;;; Internal variables:
464 (defun mm-mule-charset-to-mime-charset (charset)
465 "Return the MIME charset corresponding to the given Mule CHARSET."
466 (if (and (fboundp 'find-coding-systems-for-charsets)
467 (fboundp 'sort-coding-systems))
468 (let ((css (sort (sort-coding-systems
469 (find-coding-systems-for-charsets (list charset)))
470 'mm-sort-coding-systems-predicate))
472 (while (and (not mime)
474 (when (setq cs (pop css))
475 (setq mime (or (coding-system-get cs :mime-charset)
476 (coding-system-get cs 'mime-charset)))))
478 (let ((alist (mapcar (lambda (cs)
479 (assq cs mm-mime-mule-charset-alist))
480 (sort (mapcar 'car mm-mime-mule-charset-alist)
481 'mm-sort-coding-systems-predicate)))
484 (when (memq charset (cdar alist))
485 (setq out (caar alist)
490 (defun mm-charset-to-coding-system (charset &optional lbt
492 "Return coding-system corresponding to CHARSET.
493 CHARSET is a symbol naming a MIME charset.
494 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
495 used as the line break code type of the coding system.
497 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
498 map undesired charset names to their replacement. This should
499 only be used for decoding, not for encoding."
500 ;; OVERRIDE is used (only) in `mm-decode-body'.
501 (when (stringp charset)
502 (setq charset (intern (downcase charset))))
504 (setq charset (intern (format "%s-%s" charset lbt))))
508 ;; Running in a non-MULE environment.
509 ((or (null (mm-get-coding-system-list))
510 (not (fboundp 'coding-system-get)))
512 ;; Check override list quite early. Should only used for decoding, not for
515 (let ((cs (cdr (assq charset mm-charset-override-alist))))
516 (and cs (mm-coding-system-p cs) cs))))
518 ((eq charset 'us-ascii)
520 ;; Check to see whether we can handle this charset. (This depends
521 ;; on there being some coding system matching each `mime-charset'
522 ;; property defined, as there should be.)
523 ((and (mm-coding-system-p charset)
524 ;;; Doing this would potentially weed out incorrect charsets.
526 ;;; (eq charset (coding-system-get charset 'mime-charset))
529 ;; Eval expressions from `mm-charset-eval-alist'
530 ((let* ((el (assq charset mm-charset-eval-alist))
537 (condition-case nil (eval form) (error nil))
538 ;; (message "Failed to eval `%s'" form))
539 (mm-coding-system-p cs)
540 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
542 ;; Translate invalid charsets.
543 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
545 (mm-coding-system-p cs)
547 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
550 ;; Last resort: search the coding system list for entries which
551 ;; have the right mime-charset in case the canonical name isn't
552 ;; defined (though it should be).
554 ;; mm-get-coding-system-list returns a list of cs without lbt.
556 (dolist (c (mm-get-coding-system-list))
558 (eq charset (or (coding-system-get c :mime-charset)
559 (coding-system-get c 'mime-charset))))
562 ;; Warn the user about unknown charset:
563 (if (fboundp 'gnus-message)
564 (gnus-message 7 "Unknown charset: %s" charset)
565 (message "Unknown charset: %s" charset)))
569 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
570 (boundp 'default-enable-multibyte-characters)
571 default-enable-multibyte-characters
572 (fboundp 'set-buffer-multibyte))
573 "True in Emacs with Mule.")
576 (defun mm-enable-multibyte ()
577 "Set the multibyte flag of the current buffer.
578 Only do this if the default value of `enable-multibyte-characters' is
579 non-nil. This is a no-op in XEmacs."
580 (set-buffer-multibyte 'to))
581 (defalias 'mm-enable-multibyte 'ignore))
584 (defun mm-disable-multibyte ()
585 "Unset the multibyte flag of in the current buffer.
586 This is a no-op in XEmacs."
587 (set-buffer-multibyte nil))
588 (defalias 'mm-disable-multibyte 'ignore)))
590 (defun mm-preferred-coding-system (charset)
591 ;; A typo in some Emacs versions.
592 (or (get-charset-property charset 'preferred-coding-system)
593 (get-charset-property charset 'prefered-coding-system)))
595 ;; Mule charsets shouldn't be used.
596 (defsubst mm-guess-charset ()
597 "Guess Mule charset from the language environment."
599 mail-parse-mule-charset ;; cached mule-charset
601 (setq mail-parse-mule-charset
602 (and (boundp 'current-language-environment)
605 (assoc current-language-environment
606 language-info-alist))))))
607 (if (or (not mail-parse-mule-charset)
608 (eq mail-parse-mule-charset 'ascii))
609 (setq mail-parse-mule-charset
610 (or (car (last (assq mail-parse-charset
611 mm-mime-mule-charset-alist)))
614 mail-parse-mule-charset)))
616 (defun mm-charset-after (&optional pos)
617 "Return charset of a character in current buffer at position POS.
618 If POS is nil, it defauls to the current point.
619 If POS is out of range, the value is nil.
620 If the charset is `composition', return the actual one."
621 (let ((char (char-after pos)) charset)
622 (if (< (mm-char-int char) 128)
623 (setq charset 'ascii)
624 ;; charset-after is fake in some Emacsen.
625 (setq charset (and (fboundp 'char-charset) (char-charset char)))
626 (if (eq charset 'composition) ; Mule 4
627 (let ((p (or pos (point))))
628 (cadr (find-charset-region p (1+ p))))
629 (if (and charset (not (memq charset '(ascii eight-bit-control
630 eight-bit-graphic))))
632 (mm-guess-charset))))))
634 (defun mm-mime-charset (charset)
635 "Return the MIME charset corresponding to the given Mule CHARSET."
636 (if (eq charset 'unknown)
637 (error "The message contains non-printable characters, please use attachment"))
638 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
639 ;; This exists in Emacs 20.
641 (and (mm-preferred-coding-system charset)
642 (or (coding-system-get
643 (mm-preferred-coding-system charset) :mime-charset)
645 (mm-preferred-coding-system charset) 'mime-charset)))
646 (and (eq charset 'ascii)
648 (mm-preferred-coding-system charset)
649 (mm-mule-charset-to-mime-charset charset))
650 ;; This is for XEmacs.
651 (mm-mule-charset-to-mime-charset charset)))
653 (if (fboundp 'delete-dups)
654 (defalias 'mm-delete-duplicates 'delete-dups)
655 (defun mm-delete-duplicates (list)
656 "Destructively remove `equal' duplicates from LIST.
657 Store the result in LIST and return it. LIST must be a proper list.
658 Of several `equal' occurrences of an element in LIST, the first
661 This is a compatibility function for Emacsen without `delete-dups'."
662 ;; Code from `subr.el' in Emacs 22:
665 (setcdr tail (delete (car tail) (cdr tail)))
666 (setq tail (cdr tail))))
669 ;; Fixme: This is used in places when it should be testing the
670 ;; default multibyteness. See mm-default-multibyte-p.
672 (if (and (not (featurep 'xemacs))
673 (boundp 'enable-multibyte-characters))
674 (defun mm-multibyte-p ()
675 "Non-nil if multibyte is enabled in the current buffer."
676 enable-multibyte-characters)
677 (defun mm-multibyte-p () (featurep 'mule))))
679 (defun mm-default-multibyte-p ()
680 "Return non-nil if the session is multibyte.
681 This affects whether coding conversion should be attempted generally."
683 (if (boundp 'default-enable-multibyte-characters)
684 default-enable-multibyte-characters
687 (defun mm-iso-8859-x-to-15-region (&optional b e)
688 (if (fboundp 'char-charset)
689 (let (charset item c inconvertible)
691 (if e (narrow-to-region b e))
692 (goto-char (point-min))
693 (skip-chars-forward "\0-\177")
696 ((not (setq item (assq (char-charset (setq c (char-after)))
697 mm-iso-8859-x-to-15-table)))
699 ((memq c (cdr (cdr item)))
700 (setq inconvertible t)
703 (insert-before-markers (prog1 (+ c (car (cdr item)))
705 (skip-chars-forward "\0-\177")))
706 (not inconvertible))))
708 (defun mm-sort-coding-systems-predicate (a b)
711 ;; Note: invalid entries are dropped silently
712 (and (setq cs (mm-coding-system-p cs))
713 (coding-system-base cs)))
714 mm-coding-system-priorities)))
715 (and (setq a (mm-coding-system-p a))
716 (if (setq b (mm-coding-system-p b))
717 (> (length (memq (coding-system-base a) priorities))
718 (length (memq (coding-system-base b) priorities)))
722 (autoload 'latin-unity-massage-name "latin-unity")
723 (autoload 'latin-unity-maybe-remap "latin-unity")
724 (autoload 'latin-unity-representations-feasible-region "latin-unity")
725 (autoload 'latin-unity-representations-present-region "latin-unity")
726 (defvar latin-unity-coding-systems)
727 (defvar latin-unity-ucs-list))
729 (defun mm-xemacs-find-mime-charset-1 (begin end)
730 "Determine which MIME charset to use to send region as message.
731 This uses the XEmacs-specific latin-unity package to better handle the
732 case where identical characters from diverse ISO-8859-? character sets
733 can be encoded using a single one of the corresponding coding systems.
735 It treats `mm-coding-system-priorities' as the list of preferred
736 coding systems; a useful example setting for this list in Western
737 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
738 to the very standard Latin 1 coding system, and only move to coding
739 systems that are less supported as is necessary to encode the
740 characters that exist in the buffer.
742 Latin Unity doesn't know about those non-ASCII Roman characters that
743 are available in various East Asian character sets. As such, its
744 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
745 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
746 But this is very much a corner case, so don't worry about it."
747 (let ((systems mm-coding-system-priorities) csets psets curset)
749 ;; Load the Latin Unity library, if available.
750 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
751 (require 'latin-unity))
753 ;; Now, can we use it?
754 (if (featurep 'latin-unity)
756 (setq csets (latin-unity-representations-feasible-region begin end)
757 psets (latin-unity-representations-present-region begin end))
761 ;; Pass back the first coding system in the preferred list
762 ;; that can encode the whole region.
763 (dolist (curset systems)
764 (setq curset (latin-unity-massage-name 'buffer-default curset))
766 ;; If the coding system is a universal coding system, then
767 ;; it can certainly encode all the characters in the region.
768 (if (memq curset latin-unity-ucs-list)
769 (throw 'done (list curset)))
771 ;; If a coding system isn't universal, and isn't in
772 ;; the list that latin unity knows about, we can't
773 ;; decide whether to use it here. Leave that until later
774 ;; in `mm-find-mime-charset-region' function, whence we
776 (unless (memq curset latin-unity-coding-systems)
779 ;; Right, we know about this coding system, and it may
780 ;; conceivably be able to encode all the characters in
782 (if (latin-unity-maybe-remap begin end curset csets psets t)
783 (throw 'done (list curset))))
785 ;; Can't encode using anything from the
786 ;; `mm-coding-system-priorities' list.
787 ;; Leave `mm-find-mime-charset' to do most of the work.
790 ;; Right, latin unity isn't available; let `mm-find-charset-region'
791 ;; take its default action, which equally applies to GNU Emacs.
794 (defmacro mm-xemacs-find-mime-charset (begin end)
795 (when (featurep 'xemacs)
796 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
798 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
799 "Return the MIME charsets needed to encode the region between B and E.
800 nil means ASCII, a single-element list represents an appropriate MIME
801 charset, and a longer list means no appropriate charset."
803 ;; The return possibilities of this function are a mess...
804 (or (and (mm-multibyte-p)
805 mm-use-find-coding-systems-region
806 ;; Find the mime-charset of the most preferred coding
807 ;; system that has one.
808 (let ((systems (find-coding-systems-region b e)))
809 (when mm-coding-system-priorities
811 (sort systems 'mm-sort-coding-systems-predicate)))
812 (setq systems (delq 'compound-text systems))
813 (unless (equal systems '(undecided))
815 (let* ((head (pop systems))
816 (cs (or (coding-system-get head :mime-charset)
817 (coding-system-get head 'mime-charset))))
818 ;; The mime-charset (`x-ctext') of
819 ;; `compound-text' is not in the IANA list. We
820 ;; shouldn't normally use anything here with a
821 ;; mime-charset having an `x-' prefix.
822 ;; Fixme: Allow this to be overridden, since
823 ;; there is existing use of x-ctext.
824 ;; Also people apparently need the coding system
825 ;; `iso-2022-jp-3' (which Mule-UCS defines with
826 ;; mime-charset, though it's not valid).
828 (not (string-match "^[Xx]-" (symbol-name cs)))
829 ;; UTF-16 of any variety is invalid for
830 ;; text parts and, unfortunately, has
831 ;; mime-charset defined both in Mule-UCS
832 ;; and versions of Emacs. (The name
833 ;; might be `mule-utf-16...' or
835 (not (string-match "utf-16" (symbol-name cs))))
837 charsets (list cs))))))
839 ;; If we're XEmacs, and some coding system is appropriate,
840 ;; mm-xemacs-find-mime-charset will return an appropriate list.
841 ;; Otherwise, we'll get nil, and the next setq will get invoked.
842 (setq charsets (mm-xemacs-find-mime-charset b e))
844 ;; We're not multibyte, or a single coding system won't cover it.
846 (mm-delete-duplicates
847 (mapcar 'mm-mime-charset
849 (mm-find-charset-region b e))))))
850 (if (and (> (length charsets) 1)
851 (memq 'iso-8859-15 charsets)
852 (memq 'iso-8859-15 hack-charsets)
853 (save-excursion (mm-iso-8859-x-to-15-region b e)))
854 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
855 mm-iso-8859-15-compatible))
856 (if (and (memq 'iso-2022-jp-2 charsets)
857 (memq 'iso-2022-jp-2 hack-charsets))
858 (setq charsets (delq 'iso-2022-jp charsets)))
859 ;; Attempt to reduce the number of charsets if utf-8 is available.
860 (if (and (featurep 'xemacs)
861 (> (length charsets) 1)
862 (mm-coding-system-p 'utf-8))
863 (let ((mm-coding-system-priorities
864 (cons 'utf-8 mm-coding-system-priorities)))
866 (mm-delete-duplicates
867 (mapcar 'mm-mime-charset
869 (mm-find-charset-region b e)))))))
872 (defmacro mm-with-unibyte-buffer (&rest forms)
873 "Create a temporary buffer, and evaluate FORMS there like `progn'.
874 Use unibyte mode for this."
875 `(let (default-enable-multibyte-characters)
876 (with-temp-buffer ,@forms)))
877 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
878 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
880 (defmacro mm-with-multibyte-buffer (&rest forms)
881 "Create a temporary buffer, and evaluate FORMS there like `progn'.
882 Use multibyte mode for this."
883 `(let ((default-enable-multibyte-characters t))
884 (with-temp-buffer ,@forms)))
885 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
886 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
888 (defmacro mm-with-unibyte-current-buffer (&rest forms)
889 "Evaluate FORMS with current buffer temporarily made unibyte.
890 Also bind `default-enable-multibyte-characters' to nil.
891 Equivalent to `progn' in XEmacs"
892 (let ((multibyte (make-symbol "multibyte"))
893 (buffer (make-symbol "buffer")))
895 (let ((,multibyte enable-multibyte-characters)
896 (,buffer (current-buffer)))
898 (let (default-enable-multibyte-characters)
899 (set-buffer-multibyte nil)
902 (set-buffer-multibyte ,multibyte)))
903 (let (default-enable-multibyte-characters)
905 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
906 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
908 (defmacro mm-with-unibyte (&rest forms)
909 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
910 `(let (default-enable-multibyte-characters)
912 (put 'mm-with-unibyte 'lisp-indent-function 0)
913 (put 'mm-with-unibyte 'edebug-form-spec '(body))
915 (defmacro mm-with-multibyte (&rest forms)
916 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
917 `(let ((default-enable-multibyte-characters t))
919 (put 'mm-with-multibyte 'lisp-indent-function 0)
920 (put 'mm-with-multibyte 'edebug-form-spec '(body))
922 (defun mm-find-charset-region (b e)
923 "Return a list of Emacs charsets in the region B to E."
925 ((and (mm-multibyte-p)
926 (fboundp 'find-charset-region))
927 ;; Remove composition since the base charsets have been included.
928 ;; Remove eight-bit-*, treat them as ascii.
929 (let ((css (find-charset-region b e)))
930 (mapcar (lambda (cs) (setq css (delq cs css)))
931 '(composition eight-bit-control eight-bit-graphic
935 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
938 (narrow-to-region b e)
939 (goto-char (point-min))
940 (skip-chars-forward "\0-\177")
945 (and (boundp 'current-language-environment)
946 (car (last (assq 'charset
947 (assoc current-language-environment
948 language-info-alist))))))
949 (if (eq charset 'ascii) (setq charset nil))
952 (car (last (assq mail-parse-charset
953 mm-mime-mule-charset-alist)))))
954 (list 'ascii (or charset 'latin-iso8859-1)))))))))
956 (defun mm-auto-mode-alist ()
957 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
958 (let ((alist auto-mode-alist)
961 (when (listp (cdar alist))
962 (push (car alist) out))
966 (defvar mm-inhibit-file-name-handlers
967 '(jka-compr-handler image-file-handler)
968 "A list of handlers doing (un)compression (etc) thingies.")
970 (defun mm-insert-file-contents (filename &optional visit beg end replace
972 "Like `insert-file-contents', but only reads in the file.
973 A buffer may be modified in several ways after reading into the buffer due
974 to advanced Emacs features, such as file-name-handlers, format decoding,
975 `find-file-hooks', etc.
976 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
977 This function ensures that none of these modifications will take place."
978 (let* ((format-alist nil)
979 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
980 (default-major-mode 'fundamental-mode)
981 (enable-local-variables nil)
982 (after-insert-file-functions nil)
983 (enable-local-eval nil)
984 (inhibit-file-name-operation (if inhibit
985 'insert-file-contents
986 inhibit-file-name-operation))
987 (inhibit-file-name-handlers
989 (append mm-inhibit-file-name-handlers
990 inhibit-file-name-handlers)
991 inhibit-file-name-handlers))
992 (ffh (if (boundp 'find-file-hook)
995 (val (symbol-value ffh)))
998 (insert-file-contents filename visit beg end replace)
1001 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1002 "Append the contents of the region to the end of file FILENAME.
1003 When called from a function, expects three arguments,
1004 START, END and FILENAME. START and END are buffer positions
1005 saying what text to write.
1006 Optional fourth argument specifies the coding system to use when
1008 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1009 (let ((coding-system-for-write
1010 (or codesys mm-text-coding-system-for-write
1011 mm-text-coding-system))
1012 (inhibit-file-name-operation (if inhibit
1014 inhibit-file-name-operation))
1015 (inhibit-file-name-handlers
1017 (append mm-inhibit-file-name-handlers
1018 inhibit-file-name-handlers)
1019 inhibit-file-name-handlers)))
1020 (write-region start end filename t 'no-message)
1021 (message "Appended to %s" filename)))
1023 (defun mm-write-region (start end filename &optional append visit lockname
1024 coding-system inhibit)
1026 "Like `write-region'.
1027 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1028 (let ((coding-system-for-write
1029 (or coding-system mm-text-coding-system-for-write
1030 mm-text-coding-system))
1031 (inhibit-file-name-operation (if inhibit
1033 inhibit-file-name-operation))
1034 (inhibit-file-name-handlers
1036 (append mm-inhibit-file-name-handlers
1037 inhibit-file-name-handlers)
1038 inhibit-file-name-handlers)))
1039 (write-region start end filename append visit lockname)))
1041 (defun mm-image-load-path (&optional package)
1043 (dolist (path load-path (nreverse result))
1046 (setq dir (concat (file-name-directory
1047 (directory-file-name path))
1048 "etc/images/" (or package "gnus/")))))
1050 (push path result))))
1052 ;; Fixme: This doesn't look useful where it's used.
1053 (if (fboundp 'detect-coding-region)
1054 (defun mm-detect-coding-region (start end)
1055 "Like `detect-coding-region' except returning the best one."
1056 (let ((coding-systems
1057 (detect-coding-region start end)))
1058 (or (car-safe coding-systems)
1060 (defun mm-detect-coding-region (start end)
1061 (let ((point (point)))
1063 (skip-chars-forward "\0-\177" end)
1065 (if (eq (point) end) 'ascii (mm-guess-charset))
1066 (goto-char point)))))
1068 (if (fboundp 'coding-system-get)
1069 (defun mm-detect-mime-charset-region (start end)
1070 "Detect MIME charset of the text in the region between START and END."
1071 (let ((cs (mm-detect-coding-region start end)))
1072 (or (coding-system-get cs :mime-charset)
1073 (coding-system-get cs 'mime-charset))))
1074 (defun mm-detect-mime-charset-region (start end)
1075 "Detect MIME charset of the text in the region between START and END."
1076 (let ((cs (mm-detect-coding-region start end)))
1080 (unless (fboundp 'coding-system-to-mime-charset)
1081 (defalias 'coding-system-to-mime-charset 'ignore)))
1083 (defun mm-coding-system-to-mime-charset (coding-system)
1084 "Return the MIME charset corresponding to CODING-SYSTEM.
1085 To make this function work with XEmacs, the APEL package is required."
1087 (or (and (fboundp 'coding-system-get)
1088 (or (coding-system-get coding-system :mime-charset)
1089 (coding-system-get coding-system 'mime-charset)))
1090 (and (featurep 'xemacs)
1091 (or (and (fboundp 'coding-system-to-mime-charset)
1092 (not (eq (symbol-function 'coding-system-to-mime-charset)
1094 (and (condition-case nil
1097 (fboundp 'coding-system-to-mime-charset)))
1098 (coding-system-to-mime-charset coding-system)))))
1101 (require 'jka-compr))
1103 (defun mm-decompress-buffer (filename &optional inplace force)
1104 "Decompress buffer's contents, depending on jka-compr.
1105 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1106 agrees with `jka-compr-compression-info-list', decompression is done.
1107 Signal an error if FORCE is neither nil nor t and compressed data are
1108 not decompressed because `auto-compression-mode' is disabled.
1109 If INPLACE is nil, return decompressed data or nil without modifying
1110 the buffer. Otherwise, replace the buffer's contents with the
1111 decompressed data. The buffer's multibyteness must be turned off."
1114 (prog1 t (require 'jka-compr))
1115 (and (fboundp 'jka-compr-installed-p)
1116 (jka-compr-installed-p))))
1117 (let ((info (jka-compr-get-compression-info filename)))
1119 (unless (or (memq force (list nil t))
1120 (jka-compr-installed-p))
1122 (let ((prog (jka-compr-info-uncompress-program info))
1123 (args (jka-compr-info-uncompress-args info))
1124 (msg (format "%s %s..."
1125 (jka-compr-info-uncompress-message info)
1127 (err-file (jka-compr-make-temp-name))
1128 (cur (current-buffer))
1129 (coding-system-for-read mm-binary-coding-system)
1130 (coding-system-for-write mm-binary-coding-system)
1134 (insert-buffer-substring cur)
1137 (unless (memq (apply 'call-process-region
1138 (point-min) (point-max)
1139 prog t (list t err-file) nil args)
1140 jka-compr-acceptable-retval-list)
1144 (delete "" (split-string
1146 (insert-file-contents err-file)
1152 (format "Error while executing \"%s %s < %s\""
1153 prog (mapconcat 'identity args " ")
1155 (setq retval (buffer-string)))
1157 (setq err-msg (error-message-string err)))))
1158 (when (file-exists-p err-file)
1159 (ignore-errors (jka-compr-delete-temp-file err-file)))
1162 (delete-region (point-min) (point-max))
1165 (message "%s" (or err-msg (concat msg "done")))
1169 (unless (fboundp 'coding-system-name)
1170 (defalias 'coding-system-name 'ignore))
1171 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1172 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1173 (unless (fboundp 'find-operation-coding-system)
1174 (defalias 'find-operation-coding-system 'ignore)))
1176 (defun mm-find-buffer-file-coding-system (&optional filename)
1177 "Find coding system used to decode the contents of the current buffer.
1178 This function looks for the coding system magic cookie or examines the
1179 coding system specified by `file-coding-system-alist' being associated
1180 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1181 gzip, bzip2, etc. are allowed."
1183 (setq filename buffer-file-name))
1185 (let ((decomp (unless ;; No worth to examine charset of tar files.
1188 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1190 (mm-decompress-buffer filename nil t))))
1192 (set-buffer (let (default-enable-multibyte-characters)
1193 (generate-new-buffer " *temp*")))
1195 (setq filename (file-name-sans-extension filename)))
1196 (goto-char (point-min))
1199 ((boundp 'set-auto-coding-function) ;; Emacs
1201 (or (funcall (symbol-value 'set-auto-coding-function)
1202 filename (- (point-max) (point-min)))
1203 (car (find-operation-coding-system 'insert-file-contents
1205 (let (auto-coding-alist)
1207 (funcall (symbol-value 'set-auto-coding-function)
1208 nil (- (point-max) (point-min)))
1210 ((featurep 'file-coding) ;; XEmacs
1211 (let ((case-fold-search t)
1212 (end (point-at-eol))
1215 (and (re-search-forward "-\\*-+[\t ]*" end t)
1217 (setq start (match-end 0))
1218 (re-search-forward "[\t ]*-+\\*-" end t))
1220 (setq end (match-beginning 0))
1222 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1224 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1226 (find-coding-system (setq codesys
1227 (intern (match-string 1))))
1229 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1232 (setq start (match-end 0))
1233 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1235 (setq end (match-beginning 0))
1238 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1240 (find-coding-system (setq codesys
1241 (intern (match-string 1))))
1244 (goto-char (point-min))
1245 (setq case-fold-search nil)
1246 (re-search-forward "^;;;coding system: "
1247 ;;(+ (point-min) 3000) t))
1249 (looking-at "[^\t\n\r ]+")
1251 (setq codesys (intern (match-string 0))))
1255 (find-file-coding-system-for-read-from-filename
1257 (coding-system-name (coding-system-base codesys)))))))
1259 (kill-buffer (current-buffer)))))))
1263 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1264 ;;; mm-util.el ends here