1 ;;; mm-util.el --- Utility functions for Mule and low level things
2 ;; Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004
3 ;; Free Software Foundation, Inc.
5 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
6 ;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
7 ;; This file is part of GNU Emacs.
9 ;; GNU Emacs is free software; you can redistribute it and/or modify
10 ;; it under the terms of the GNU General Public License as published by
11 ;; the Free Software Foundation; either version 2, or (at your option)
14 ;; GNU Emacs is distributed in the hope that it will be useful,
15 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
16 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 ;; GNU General Public License for more details.
19 ;; You should have received a copy of the GNU General Public License
20 ;; along with GNU Emacs; see the file COPYING. If not, write to the
21 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
22 ;; Boston, MA 02111-1307, USA.
28 (eval-when-compile (require 'cl))
34 (let ((nfunc (intern (format "mm-%s" (car elem)))))
35 (if (fboundp (car elem))
36 (defalias nfunc (car elem))
37 (defalias nfunc (cdr elem)))))
38 '((decode-coding-string . (lambda (s a) s))
39 (encode-coding-string . (lambda (s a) s))
40 (encode-coding-region . ignore)
41 (coding-system-list . ignore)
42 (decode-coding-region . ignore)
44 (coding-system-equal . equal)
45 (annotationp . ignore)
46 (set-buffer-file-coding-system . ignore)
53 (mapcar (lambda (e) (list (symbol-name (car e))))
54 mm-mime-mule-charset-alist)
57 . (lambda (from to string &optional inplace) ;; stolen (and renamed) from nnheader.el
58 "Replace characters in STRING from FROM to TO.
59 Unless optional argument INPLACE is non-nil, return a new string."
60 (let ((string (if inplace string (copy-sequence string)))
63 ;; Replace all occurrences of FROM with TO.
65 (when (= (aref string idx) from)
70 . (lambda (string regexp rep &optional literal)
71 "See `replace-regexp-in-string', only the order of args differs."
72 (replace-regexp-in-string regexp rep string nil literal)))
73 (string-as-unibyte . identity)
74 (string-make-unibyte . identity)
75 (string-as-multibyte . identity)
76 (multibyte-string-p . ignore)
77 ;; It is not a MIME function, but some MIME functions use it.
78 (make-temp-file . (lambda (prefix &optional dir-flag)
79 (let ((file (expand-file-name
80 (make-temp-name prefix)
81 (if (fboundp 'temp-directory)
83 temporary-file-directory))))
85 (make-directory file))
87 (insert-byte . insert-char)
88 (multibyte-char-to-unibyte . identity))))
91 (defalias 'mm-char-or-char-int-p
93 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
94 ((fboundp 'char-valid-p) 'char-valid-p)
97 ;; Fixme: This seems always to be used to read a MIME charset, so it
98 ;; should be re-named and fixed (in Emacs) to offer completion only on
99 ;; proper charset names (base coding systems which have a
100 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
102 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
103 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
104 ;; Actually, there should be an `mm-coding-system-mime-charset'.
106 (defalias 'mm-read-coding-system
108 ((fboundp 'read-coding-system)
109 (if (and (featurep 'xemacs)
110 (<= (string-to-number emacs-version) 21.1))
111 (lambda (prompt &optional default-coding-system)
112 (read-coding-system prompt))
113 'read-coding-system))
114 (t (lambda (prompt &optional default-coding-system)
115 "Prompt the user for a coding system."
117 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
118 mm-mime-mule-charset-alist)))))))
120 (defvar mm-coding-system-list nil)
121 (defun mm-get-coding-system-list ()
122 "Get the coding system list."
123 (or mm-coding-system-list
124 (setq mm-coding-system-list (mm-coding-system-list))))
126 (defun mm-coding-system-p (cs)
127 "Return non-nil if CS is a symbol naming a coding system.
128 In XEmacs, also return non-nil if CS is a coding system object.
129 If CS is available, return CS itself in Emacs, and return a coding
130 system object in XEmacs."
131 (if (fboundp 'find-coding-system)
132 (find-coding-system cs)
133 (if (fboundp 'coding-system-p)
134 (when (coding-system-p cs)
136 ;; Is this branch ever actually useful?
137 (car (memq cs (mm-get-coding-system-list))))))
139 (defvar mm-charset-synonym-alist
141 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
142 ,@(unless (mm-coding-system-p 'x-ctext)
143 '((x-ctext . ctext)))
144 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_!
145 ,@(unless (mm-coding-system-p 'iso-8859-15)
146 '((iso-8859-15 . iso-8859-1)))
147 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
148 ,@(unless (mm-coding-system-p 'big5-hkscs)
149 '((big5-hkscs . big5)))
150 ;; Windows-1252 is actually a superset of Latin-1. See also
151 ;; `gnus-article-dumbquotes-map'.
152 ,@(unless (mm-coding-system-p 'windows-1252)
153 (if (mm-coding-system-p 'cp1252)
154 '((windows-1252 . cp1252))
155 '((windows-1252 . iso-8859-1))))
156 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
157 ;; Outlook users in Czech republic. Use this to allow reading of their
158 ;; e-mails. cp1250 should be defined by M-x codepage-setup.
159 ,@(if (and (not (mm-coding-system-p 'windows-1250))
160 (mm-coding-system-p 'cp1250))
161 '((windows-1250 . cp1250)))
162 ;; A Microsoft misunderstanding.
163 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
164 (if (mm-coding-system-p 'cp949)
165 '((ks_c_5601-1987 . cp949))
166 '((ks_c_5601-1987 . euc-kr))))
168 "A mapping from invalid charset names to the real charset names.")
170 (defvar mm-binary-coding-system
172 ((mm-coding-system-p 'binary) 'binary)
173 ((mm-coding-system-p 'no-conversion) 'no-conversion)
175 "100% binary coding system.")
177 (defvar mm-text-coding-system
178 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
179 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
180 (and (mm-coding-system-p 'raw-text) 'raw-text))
181 mm-binary-coding-system)
182 "Text-safe coding system (For removing ^M).")
184 (defvar mm-text-coding-system-for-write nil
185 "Text coding system for write.")
187 (defvar mm-auto-save-coding-system
189 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
190 (if (memq system-type '(windows-nt ms-dos ms-windows))
191 (if (mm-coding-system-p 'utf-8-emacs-dos)
192 'utf-8-emacs-dos mm-binary-coding-system)
194 ((mm-coding-system-p 'emacs-mule)
195 (if (memq system-type '(windows-nt ms-dos ms-windows))
196 (if (mm-coding-system-p 'emacs-mule-dos)
197 'emacs-mule-dos mm-binary-coding-system)
199 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
200 (t mm-binary-coding-system))
201 "Coding system of auto save file.")
203 (defvar mm-universal-coding-system mm-auto-save-coding-system
204 "The universal coding system.")
206 ;; Fixme: some of the cars here aren't valid MIME charsets. That
207 ;; should only matter with XEmacs, though.
208 (defvar mm-mime-mule-charset-alist
210 (iso-8859-1 latin-iso8859-1)
211 (iso-8859-2 latin-iso8859-2)
212 (iso-8859-3 latin-iso8859-3)
213 (iso-8859-4 latin-iso8859-4)
214 (iso-8859-5 cyrillic-iso8859-5)
215 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
216 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
217 ;; charset is koi8-r, not iso-8859-5.
218 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
219 (iso-8859-6 arabic-iso8859-6)
220 (iso-8859-7 greek-iso8859-7)
221 (iso-8859-8 hebrew-iso8859-8)
222 (iso-8859-9 latin-iso8859-9)
223 (iso-8859-14 latin-iso8859-14)
224 (iso-8859-15 latin-iso8859-15)
225 (viscii vietnamese-viscii-lower)
226 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
227 (euc-kr korean-ksc5601)
228 (gb2312 chinese-gb2312)
229 (big5 chinese-big5-1 chinese-big5-2)
231 (thai-tis620 thai-tis620)
232 (windows-1251 cyrillic-iso8859-5)
233 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
234 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
235 latin-jisx0201 japanese-jisx0208-1978
236 chinese-gb2312 japanese-jisx0208
237 korean-ksc5601 japanese-jisx0212)
238 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
239 latin-jisx0201 japanese-jisx0208-1978
240 chinese-gb2312 japanese-jisx0208
241 korean-ksc5601 japanese-jisx0212
242 chinese-cns11643-1 chinese-cns11643-2)
243 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
244 cyrillic-iso8859-5 greek-iso8859-7
245 latin-jisx0201 japanese-jisx0208-1978
246 chinese-gb2312 japanese-jisx0208
247 korean-ksc5601 japanese-jisx0212
248 chinese-cns11643-1 chinese-cns11643-2
249 chinese-cns11643-3 chinese-cns11643-4
250 chinese-cns11643-5 chinese-cns11643-6
252 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
253 japanese-jisx0213-1 japanese-jisx0213-2)
254 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
255 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
256 (charsetp 'unicode-a)
257 (not (mm-coding-system-p 'mule-utf-8)))
258 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
259 ;; If we have utf-8 we're in Mule 5+.
262 (coding-system-get 'mule-utf-8 'safe-charsets)))))
263 "Alist of MIME-charset/MULE-charsets.")
265 (defun mm-enrich-utf-8-by-mule-ucs ()
266 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
267 This function will run when the `un-define' module is loaded under
268 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
269 with Mule charsets. It is completely useless for Emacs."
270 (setq after-load-alist
271 (delete '("un-define" (mm-enrich-utf-8-by-mule-ucs))
273 (when (boundp 'unicode-basic-translation-charset-order-list)
278 'unicode-basic-translation-charset-order-list)))
279 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
282 (setq mm-mime-mule-charset-alist
283 (nconc mm-mime-mule-charset-alist
284 (list (cons 'utf-8 val))))))
287 ;; Correct by construction, but should be unnecessary for Emacs:
288 (if (featurep 'xemacs)
289 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
290 (when (and (fboundp 'coding-system-list)
291 (fboundp 'sort-coding-systems))
292 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
296 mime (or (coding-system-get cs :mime-charset) ; Emacs 22
297 (coding-system-get cs 'mime-charset)))
299 (not (eq t (setq mule
300 (coding-system-get cs 'safe-charsets))))
301 (not (assq mime alist)))
302 (push (cons mime (delq 'ascii mule)) alist)))
303 (setq mm-mime-mule-charset-alist (nreverse alist)))))
305 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
306 "A list of special charsets.
307 Valid elements include:
308 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
309 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
312 (defvar mm-iso-8859-15-compatible
313 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
314 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
315 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
317 (defvar mm-iso-8859-x-to-15-table
318 (and (fboundp 'coding-system-p)
319 (mm-coding-system-p 'iso-8859-15)
322 (if (mm-coding-system-p (car cs))
323 (let ((c (string-to-char
324 (decode-coding-string "\341" (car cs)))))
325 (cons (char-charset c)
328 (decode-coding-string "\341" 'iso-8859-15)) c)
329 (string-to-list (decode-coding-string (car (cdr cs))
332 mm-iso-8859-15-compatible))
333 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
335 (defcustom mm-coding-system-priorities
336 (if (boundp 'current-language-environment)
337 (let ((lang (symbol-value 'current-language-environment)))
338 (cond ((string= lang "Japanese")
339 ;; Japanese users may prefer iso-2022-jp to shift_jis.
340 '(iso-2022-jp iso-2022-jp-2 shift_jis iso-8859-1 utf-8)))))
341 "Preferred coding systems for encoding outgoing messages.
343 More than one suitable coding system may be found for some text.
344 By default, the coding system with the highest priority is used
345 to encode outgoing messages (see `sort-coding-systems'). If this
346 variable is set, it overrides the default priority."
347 :type '(repeat (symbol :tag "Coding system"))
351 (defvar mm-use-find-coding-systems-region
352 (fboundp 'find-coding-systems-region)
353 "Use `find-coding-systems-region' to find proper coding systems.
355 Setting it to nil is useful on Emacsen supporting Unicode if sending
356 mail with multiple parts is preferred to sending a Unicode one.")
358 ;;; Internal variables:
362 (defun mm-mule-charset-to-mime-charset (charset)
363 "Return the MIME charset corresponding to the given Mule CHARSET."
364 (if (and (fboundp 'find-coding-systems-for-charsets)
365 (fboundp 'sort-coding-systems))
366 (let ((css (sort (sort-coding-systems
367 (find-coding-systems-for-charsets (list charset)))
368 'mm-sort-coding-systems-predicate))
370 (while (and (not mime)
372 (when (setq cs (pop css))
373 (setq mime (or (coding-system-get cs :mime-charset)
374 (coding-system-get cs 'mime-charset)))))
376 (let ((alist (mapcar (lambda (cs)
377 (assq cs mm-mime-mule-charset-alist))
378 (sort (mapcar 'car mm-mime-mule-charset-alist)
379 'mm-sort-coding-systems-predicate)))
382 (when (memq charset (cdar alist))
383 (setq out (caar alist)
388 (defun mm-charset-to-coding-system (charset &optional lbt)
389 "Return coding-system corresponding to CHARSET.
390 CHARSET is a symbol naming a MIME charset.
391 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
392 used as the line break code type of the coding system."
393 (when (stringp charset)
394 (setq charset (intern (downcase charset))))
396 (setq charset (intern (format "%s-%s" charset lbt))))
400 ;; Running in a non-MULE environment.
401 ((or (null (mm-get-coding-system-list))
402 (not (fboundp 'coding-system-get)))
405 ((eq charset 'us-ascii)
407 ;; Check to see whether we can handle this charset. (This depends
408 ;; on there being some coding system matching each `mime-charset'
409 ;; property defined, as there should be.)
410 ((and (mm-coding-system-p charset)
411 ;;; Doing this would potentially weed out incorrect charsets.
413 ;;; (eq charset (coding-system-get charset 'mime-charset))
416 ;; Translate invalid charsets.
417 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
418 (and cs (mm-coding-system-p cs) cs)))
419 ;; Last resort: search the coding system list for entries which
420 ;; have the right mime-charset in case the canonical name isn't
421 ;; defined (though it should be).
423 ;; mm-get-coding-system-list returns a list of cs without lbt.
425 (dolist (c (mm-get-coding-system-list))
427 (eq charset (or (coding-system-get c :mime-charset)
428 (coding-system-get c 'mime-charset))))
433 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
434 (boundp 'default-enable-multibyte-characters)
435 default-enable-multibyte-characters
436 (fboundp 'set-buffer-multibyte))
437 "True in Emacs with Mule.")
440 (defun mm-enable-multibyte ()
441 "Set the multibyte flag of the current buffer.
442 Only do this if the default value of `enable-multibyte-characters' is
443 non-nil. This is a no-op in XEmacs."
444 (set-buffer-multibyte 'to))
445 (defalias 'mm-enable-multibyte 'ignore))
448 (defun mm-disable-multibyte ()
449 "Unset the multibyte flag of in the current buffer.
450 This is a no-op in XEmacs."
451 (set-buffer-multibyte nil))
452 (defalias 'mm-disable-multibyte 'ignore)))
454 (defun mm-preferred-coding-system (charset)
455 ;; A typo in some Emacs versions.
456 (or (get-charset-property charset 'preferred-coding-system)
457 (get-charset-property charset 'prefered-coding-system)))
459 ;; Mule charsets shouldn't be used.
460 (defsubst mm-guess-charset ()
461 "Guess Mule charset from the language environment."
463 mail-parse-mule-charset ;; cached mule-charset
465 (setq mail-parse-mule-charset
466 (and (boundp 'current-language-environment)
469 (assoc current-language-environment
470 language-info-alist))))))
471 (if (or (not mail-parse-mule-charset)
472 (eq mail-parse-mule-charset 'ascii))
473 (setq mail-parse-mule-charset
474 (or (car (last (assq mail-parse-charset
475 mm-mime-mule-charset-alist)))
478 mail-parse-mule-charset)))
480 (defun mm-charset-after (&optional pos)
481 "Return charset of a character in current buffer at position POS.
482 If POS is nil, it defauls to the current point.
483 If POS is out of range, the value is nil.
484 If the charset is `composition', return the actual one."
485 (let ((char (char-after pos)) charset)
486 (if (< (mm-char-int char) 128)
487 (setq charset 'ascii)
488 ;; charset-after is fake in some Emacsen.
489 (setq charset (and (fboundp 'char-charset) (char-charset char)))
490 (if (eq charset 'composition) ; Mule 4
491 (let ((p (or pos (point))))
492 (cadr (find-charset-region p (1+ p))))
493 (if (and charset (not (memq charset '(ascii eight-bit-control
494 eight-bit-graphic))))
496 (mm-guess-charset))))))
498 (defun mm-mime-charset (charset)
499 "Return the MIME charset corresponding to the given Mule CHARSET."
500 (if (eq charset 'unknown)
501 (error "The message contains non-printable characters, please use attachment"))
502 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
503 ;; This exists in Emacs 20.
505 (and (mm-preferred-coding-system charset)
506 (or (coding-system-get
507 (mm-preferred-coding-system charset) :mime-charset)
509 (mm-preferred-coding-system charset) 'mime-charset)))
510 (and (eq charset 'ascii)
512 (mm-preferred-coding-system charset)
513 (mm-mule-charset-to-mime-charset charset))
514 ;; This is for XEmacs.
515 (mm-mule-charset-to-mime-charset charset)))
517 (defun mm-delete-duplicates (list)
518 "Simple substitute for CL `delete-duplicates', testing with `equal'."
521 (setq head (car list))
522 (setq list (delete head list))
523 (setq result (cons head result)))
526 ;; Fixme: This is used in places when it should be testing the
527 ;; default multibyteness. See mm-default-multibyte-p.
529 (if (and (not (featurep 'xemacs))
530 (boundp 'enable-multibyte-characters))
531 (defun mm-multibyte-p ()
532 "Non-nil if multibyte is enabled in the current buffer."
533 enable-multibyte-characters)
534 (defun mm-multibyte-p () (featurep 'mule))))
536 (defun mm-default-multibyte-p ()
537 "Return non-nil if the session is multibyte.
538 This affects whether coding conversion should be attempted generally."
540 (if (boundp 'default-enable-multibyte-characters)
541 default-enable-multibyte-characters
544 (defun mm-iso-8859-x-to-15-region (&optional b e)
545 (if (fboundp 'char-charset)
546 (let (charset item c inconvertible)
548 (if e (narrow-to-region b e))
549 (goto-char (point-min))
550 (skip-chars-forward "\0-\177")
553 ((not (setq item (assq (char-charset (setq c (char-after)))
554 mm-iso-8859-x-to-15-table)))
556 ((memq c (cdr (cdr item)))
557 (setq inconvertible t)
560 (insert-before-markers (prog1 (+ c (car (cdr item)))
562 (skip-chars-forward "\0-\177")))
563 (not inconvertible))))
565 (defun mm-sort-coding-systems-predicate (a b)
568 ;; Note: invalid entries are dropped silently
569 (and (setq cs (mm-coding-system-p cs))
570 (coding-system-base cs)))
571 mm-coding-system-priorities)))
572 (and (setq a (mm-coding-system-p a))
573 (if (setq b (mm-coding-system-p b))
574 (> (length (memq (coding-system-base a) priorities))
575 (length (memq (coding-system-base b) priorities)))
578 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
579 "Return the MIME charsets needed to encode the region between B and E.
580 nil means ASCII, a single-element list represents an appropriate MIME
581 charset, and a longer list means no appropriate charset."
583 ;; The return possibilities of this function are a mess...
584 (or (and (mm-multibyte-p)
585 mm-use-find-coding-systems-region
586 ;; Find the mime-charset of the most preferred coding
587 ;; system that has one.
588 (let ((systems (find-coding-systems-region b e)))
589 (when mm-coding-system-priorities
591 (sort systems 'mm-sort-coding-systems-predicate)))
592 (setq systems (delq 'compound-text systems))
593 (unless (equal systems '(undecided))
595 (let* ((head (pop systems))
596 (cs (or (coding-system-get head :mime-charset)
597 (coding-system-get head 'mime-charset))))
598 ;; The mime-charset (`x-ctext') of
599 ;; `compound-text' is not in the IANA list. We
600 ;; shouldn't normally use anything here with a
601 ;; mime-charset having an `x-' prefix.
602 ;; Fixme: Allow this to be overridden, since
603 ;; there is existing use of x-ctext.
604 ;; Also people apparently need the coding system
605 ;; `iso-2022-jp-3' (which Mule-UCS defines with
606 ;; mime-charset, though it's not valid).
608 (not (string-match "^[Xx]-" (symbol-name cs)))
609 ;; UTF-16 of any variety is invalid for
610 ;; text parts and, unfortunately, has
611 ;; mime-charset defined both in Mule-UCS
612 ;; and versions of Emacs. (The name
613 ;; might be `mule-utf-16...' or
615 (not (string-match "utf-16" (symbol-name cs))))
617 charsets (list cs))))))
619 ;; Otherwise we're not multibyte, we're XEmacs, or a single
620 ;; coding system won't cover it.
622 (mm-delete-duplicates
623 (mapcar 'mm-mime-charset
625 (mm-find-charset-region b e))))))
626 (if (and (> (length charsets) 1)
627 (memq 'iso-8859-15 charsets)
628 (memq 'iso-8859-15 hack-charsets)
629 (save-excursion (mm-iso-8859-x-to-15-region b e)))
630 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
631 mm-iso-8859-15-compatible))
632 (if (and (memq 'iso-2022-jp-2 charsets)
633 (memq 'iso-2022-jp-2 hack-charsets))
634 (setq charsets (delq 'iso-2022-jp charsets)))
637 (defmacro mm-with-unibyte-buffer (&rest forms)
638 "Create a temporary buffer, and evaluate FORMS there like `progn'.
639 Use unibyte mode for this."
640 `(let (default-enable-multibyte-characters)
641 (with-temp-buffer ,@forms)))
642 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
643 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
645 (defmacro mm-with-multibyte-buffer (&rest forms)
646 "Create a temporary buffer, and evaluate FORMS there like `progn'.
647 Use multibyte mode for this."
648 `(let ((default-enable-multibyte-characters t))
649 (with-temp-buffer ,@forms)))
650 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
651 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
653 (defmacro mm-with-unibyte-current-buffer (&rest forms)
654 "Evaluate FORMS with current buffer temporarily made unibyte.
655 Also bind `default-enable-multibyte-characters' to nil.
656 Equivalent to `progn' in XEmacs"
657 (let ((multibyte (make-symbol "multibyte"))
658 (buffer (make-symbol "buffer")))
660 (let ((,multibyte enable-multibyte-characters)
661 (,buffer (current-buffer)))
663 (let (default-enable-multibyte-characters)
664 (set-buffer-multibyte nil)
667 (set-buffer-multibyte ,multibyte)))
668 (let (default-enable-multibyte-characters)
670 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
671 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
673 (defmacro mm-with-unibyte (&rest forms)
674 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
675 `(let (default-enable-multibyte-characters)
677 (put 'mm-with-unibyte 'lisp-indent-function 0)
678 (put 'mm-with-unibyte 'edebug-form-spec '(body))
680 (defmacro mm-with-multibyte (&rest forms)
681 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
682 `(let ((default-enable-multibyte-characters t))
684 (put 'mm-with-multibyte 'lisp-indent-function 0)
685 (put 'mm-with-multibyte 'edebug-form-spec '(body))
687 (defun mm-find-charset-region (b e)
688 "Return a list of Emacs charsets in the region B to E."
690 ((and (mm-multibyte-p)
691 (fboundp 'find-charset-region))
692 ;; Remove composition since the base charsets have been included.
693 ;; Remove eight-bit-*, treat them as ascii.
694 (let ((css (find-charset-region b e)))
695 (mapcar (lambda (cs) (setq css (delq cs css)))
696 '(composition eight-bit-control eight-bit-graphic
700 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
703 (narrow-to-region b e)
704 (goto-char (point-min))
705 (skip-chars-forward "\0-\177")
710 (and (boundp 'current-language-environment)
711 (car (last (assq 'charset
712 (assoc current-language-environment
713 language-info-alist))))))
714 (if (eq charset 'ascii) (setq charset nil))
717 (car (last (assq mail-parse-charset
718 mm-mime-mule-charset-alist)))))
719 (list 'ascii (or charset 'latin-iso8859-1)))))))))
721 (defun mm-auto-mode-alist ()
722 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
723 (let ((alist auto-mode-alist)
726 (when (listp (cdar alist))
727 (push (car alist) out))
731 (defvar mm-inhibit-file-name-handlers
732 '(jka-compr-handler image-file-handler)
733 "A list of handlers doing (un)compression (etc) thingies.")
735 (defun mm-insert-file-contents (filename &optional visit beg end replace
737 "Like `insert-file-contents', but only reads in the file.
738 A buffer may be modified in several ways after reading into the buffer due
739 to advanced Emacs features, such as file-name-handlers, format decoding,
740 `find-file-hooks', etc.
741 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
742 This function ensures that none of these modifications will take place."
743 (let ((format-alist nil)
744 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
745 (default-major-mode 'fundamental-mode)
746 (enable-local-variables nil)
747 (after-insert-file-functions nil)
748 (enable-local-eval nil)
749 (find-file-hooks nil)
750 (inhibit-file-name-operation (if inhibit
751 'insert-file-contents
752 inhibit-file-name-operation))
753 (inhibit-file-name-handlers
755 (append mm-inhibit-file-name-handlers
756 inhibit-file-name-handlers)
757 inhibit-file-name-handlers)))
758 (insert-file-contents filename visit beg end replace)))
760 (defun mm-append-to-file (start end filename &optional codesys inhibit)
761 "Append the contents of the region to the end of file FILENAME.
762 When called from a function, expects three arguments,
763 START, END and FILENAME. START and END are buffer positions
764 saying what text to write.
765 Optional fourth argument specifies the coding system to use when
767 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
768 (let ((coding-system-for-write
769 (or codesys mm-text-coding-system-for-write
770 mm-text-coding-system))
771 (inhibit-file-name-operation (if inhibit
773 inhibit-file-name-operation))
774 (inhibit-file-name-handlers
776 (append mm-inhibit-file-name-handlers
777 inhibit-file-name-handlers)
778 inhibit-file-name-handlers)))
779 (write-region start end filename t 'no-message)
780 (message "Appended to %s" filename)))
782 (defun mm-write-region (start end filename &optional append visit lockname
783 coding-system inhibit)
785 "Like `write-region'.
786 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
787 (let ((coding-system-for-write
788 (or coding-system mm-text-coding-system-for-write
789 mm-text-coding-system))
790 (inhibit-file-name-operation (if inhibit
792 inhibit-file-name-operation))
793 (inhibit-file-name-handlers
795 (append mm-inhibit-file-name-handlers
796 inhibit-file-name-handlers)
797 inhibit-file-name-handlers)))
798 (write-region start end filename append visit lockname)))
800 (defun mm-image-load-path (&optional package)
802 (dolist (path load-path (nreverse result))
805 (setq dir (concat (file-name-directory
806 (directory-file-name path))
807 "etc/" (or package "gnus/")))))
809 (push path result))))
811 ;; Fixme: This doesn't look useful where it's used.
812 (if (fboundp 'detect-coding-region)
813 (defun mm-detect-coding-region (start end)
814 "Like `detect-coding-region' except returning the best one."
815 (let ((coding-systems
816 (detect-coding-region (point) (point-max))))
817 (or (car-safe coding-systems)
819 (defun mm-detect-coding-region (start end)
820 (let ((point (point)))
822 (skip-chars-forward "\0-\177" end)
824 (if (eq (point) end) 'ascii (mm-guess-charset))
825 (goto-char point)))))
827 (if (fboundp 'coding-system-get)
828 (defun mm-detect-mime-charset-region (start end)
829 "Detect MIME charset of the text in the region between START and END."
830 (let ((cs (mm-detect-coding-region start end)))
831 (coding-system-get cs 'mime-charset)))
832 (defun mm-detect-mime-charset-region (start end)
833 "Detect MIME charset of the text in the region between START and END."
834 (let ((cs (mm-detect-coding-region start end)))
840 ;;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
841 ;;; mm-util.el ends here