(spam-stat-unload-hook): Set as a variable w/ add-hook.
[bpt/emacs.git] / lisp / gnus / mm-util.el
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.
4
5 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
6 ;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
7 ;; This file is part of GNU Emacs.
8
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)
12 ;; any later version.
13
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.
18
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.
23
24 ;;; Commentary:
25
26 ;;; Code:
27
28 (eval-when-compile (require 'cl))
29 (require 'mail-prsvr)
30
31 (eval-and-compile
32 (mapcar
33 (lambda (elem)
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)
43 (char-int . identity)
44 (coding-system-equal . equal)
45 (annotationp . ignore)
46 (set-buffer-file-coding-system . ignore)
47 (make-char
48 . (lambda (charset int)
49 (int-to-char int)))
50 (read-charset
51 . (lambda (prompt)
52 "Return a charset."
53 (intern
54 (completing-read
55 prompt
56 (mapcar (lambda (e) (list (symbol-name (car e))))
57 mm-mime-mule-charset-alist)
58 nil t))))
59 (subst-char-in-string
60 . (lambda (from to string) ;; stolen (and renamed) from nnheader.el
61 "Replace characters in STRING from FROM to TO."
62 (let ((string (substring string 0)) ;Copy string.
63 (len (length string))
64 (idx 0))
65 ;; Replace all occurrences of FROM with TO.
66 (while (< idx len)
67 (when (= (aref string idx) from)
68 (aset string idx to))
69 (setq idx (1+ idx)))
70 string)))
71 (string-as-unibyte . identity)
72 (string-make-unibyte . identity)
73 (string-as-multibyte . identity)
74 (multibyte-string-p . ignore)
75 ;; It is not a MIME function, but some MIME functions use it.
76 (make-temp-file . (lambda (prefix &optional dir-flag)
77 (let ((file (expand-file-name
78 (make-temp-name prefix)
79 (if (fboundp 'temp-directory)
80 (temp-directory)
81 temporary-file-directory))))
82 (if dir-flag
83 (make-directory file))
84 file)))
85 (insert-byte . insert-char)
86 (multibyte-char-to-unibyte . identity))))
87
88 (eval-and-compile
89 (defalias 'mm-char-or-char-int-p
90 (cond
91 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
92 ((fboundp 'char-valid-p) 'char-valid-p)
93 (t 'identity))))
94
95 ;; Fixme: This seems always to be used to read a MIME charset, so it
96 ;; should be re-named and fixed (in Emacs) to offer completion only on
97 ;; proper charset names (base coding systems which have a
98 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
99 ;; test with
100 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
101 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
102 ;; Actually, there should be an `mm-coding-system-mime-charset'.
103 (eval-and-compile
104 (defalias 'mm-read-coding-system
105 (cond
106 ((fboundp 'read-coding-system)
107 (if (and (featurep 'xemacs)
108 (<= (string-to-number emacs-version) 21.1))
109 (lambda (prompt &optional default-coding-system)
110 (read-coding-system prompt))
111 'read-coding-system))
112 (t (lambda (prompt &optional default-coding-system)
113 "Prompt the user for a coding system."
114 (completing-read
115 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
116 mm-mime-mule-charset-alist)))))))
117
118 (defvar mm-coding-system-list nil)
119 (defun mm-get-coding-system-list ()
120 "Get the coding system list."
121 (or mm-coding-system-list
122 (setq mm-coding-system-list (mm-coding-system-list))))
123
124 (defun mm-coding-system-p (cs)
125 "Return non-nil if CS is a symbol naming a coding system.
126 In XEmacs, also return non-nil if CS is a coding system object."
127 (if (fboundp 'find-coding-system)
128 (find-coding-system cs)
129 (if (fboundp 'coding-system-p)
130 (coding-system-p cs)
131 ;; Is this branch ever actually useful?
132 (memq cs (mm-get-coding-system-list)))))
133
134 (defvar mm-charset-synonym-alist
135 `(
136 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
137 ,@(unless (mm-coding-system-p 'x-ctext)
138 '((x-ctext . ctext)))
139 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_!
140 ,@(unless (mm-coding-system-p 'iso-8859-15)
141 '((iso-8859-15 . iso-8859-1)))
142 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
143 ,@(unless (mm-coding-system-p 'big5-hkscs)
144 '((big5-hkscs . big5)))
145 ;; Windows-1252 is actually a superset of Latin-1. See also
146 ;; `gnus-article-dumbquotes-map'.
147 ,@(unless (mm-coding-system-p 'windows-1252)
148 (if (mm-coding-system-p 'cp1252)
149 '((windows-1252 . cp1252))
150 '((windows-1252 . iso-8859-1))))
151 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
152 ;; Outlook users in Czech republic. Use this to allow reading of their
153 ;; e-mails. cp1250 should be defined by M-x codepage-setup.
154 ,@(if (and (not (mm-coding-system-p 'windows-1250))
155 (mm-coding-system-p 'cp1250))
156 '((windows-1250 . cp1250)))
157 )
158 "A mapping from invalid charset names to the real charset names.")
159
160 (defvar mm-binary-coding-system
161 (cond
162 ((mm-coding-system-p 'binary) 'binary)
163 ((mm-coding-system-p 'no-conversion) 'no-conversion)
164 (t nil))
165 "100% binary coding system.")
166
167 (defvar mm-text-coding-system
168 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
169 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
170 (and (mm-coding-system-p 'raw-text) 'raw-text))
171 mm-binary-coding-system)
172 "Text-safe coding system (For removing ^M).")
173
174 (defvar mm-text-coding-system-for-write nil
175 "Text coding system for write.")
176
177 (defvar mm-auto-save-coding-system
178 (cond
179 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
180 (if (memq system-type '(windows-nt ms-dos ms-windows))
181 (if (mm-coding-system-p 'utf-8-emacs-dos)
182 'utf-8-emacs-dos mm-binary-coding-system)
183 'utf-8-emacs))
184 ((mm-coding-system-p 'emacs-mule)
185 (if (memq system-type '(windows-nt ms-dos ms-windows))
186 (if (mm-coding-system-p 'emacs-mule-dos)
187 'emacs-mule-dos mm-binary-coding-system)
188 'emacs-mule))
189 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
190 (t mm-binary-coding-system))
191 "Coding system of auto save file.")
192
193 (defvar mm-universal-coding-system mm-auto-save-coding-system
194 "The universal coding system.")
195
196 ;; Fixme: some of the cars here aren't valid MIME charsets. That
197 ;; should only matter with XEmacs, though.
198 (defvar mm-mime-mule-charset-alist
199 `((us-ascii ascii)
200 (iso-8859-1 latin-iso8859-1)
201 (iso-8859-2 latin-iso8859-2)
202 (iso-8859-3 latin-iso8859-3)
203 (iso-8859-4 latin-iso8859-4)
204 (iso-8859-5 cyrillic-iso8859-5)
205 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
206 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
207 ;; charset is koi8-r, not iso-8859-5.
208 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
209 (iso-8859-6 arabic-iso8859-6)
210 (iso-8859-7 greek-iso8859-7)
211 (iso-8859-8 hebrew-iso8859-8)
212 (iso-8859-9 latin-iso8859-9)
213 (iso-8859-14 latin-iso8859-14)
214 (iso-8859-15 latin-iso8859-15)
215 (viscii vietnamese-viscii-lower)
216 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
217 (euc-kr korean-ksc5601)
218 (gb2312 chinese-gb2312)
219 (big5 chinese-big5-1 chinese-big5-2)
220 (tibetan tibetan)
221 (thai-tis620 thai-tis620)
222 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
223 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
224 latin-jisx0201 japanese-jisx0208-1978
225 chinese-gb2312 japanese-jisx0208
226 korean-ksc5601 japanese-jisx0212
227 katakana-jisx0201)
228 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
229 latin-jisx0201 japanese-jisx0208-1978
230 chinese-gb2312 japanese-jisx0208
231 korean-ksc5601 japanese-jisx0212
232 chinese-cns11643-1 chinese-cns11643-2)
233 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
234 cyrillic-iso8859-5 greek-iso8859-7
235 latin-jisx0201 japanese-jisx0208-1978
236 chinese-gb2312 japanese-jisx0208
237 korean-ksc5601 japanese-jisx0212
238 chinese-cns11643-1 chinese-cns11643-2
239 chinese-cns11643-3 chinese-cns11643-4
240 chinese-cns11643-5 chinese-cns11643-6
241 chinese-cns11643-7)
242 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
243 (charsetp 'unicode-a)
244 (not (mm-coding-system-p 'mule-utf-8)))
245 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
246 ;; If we have utf-8 we're in Mule 5+.
247 (append '(utf-8)
248 (delete 'ascii
249 (coding-system-get 'mule-utf-8 'safe-charsets)))))
250 "Alist of MIME-charset/MULE-charsets.")
251
252 ;; Correct by construction, but should be unnecessary:
253 ;; XEmacs hates it.
254 (when (and (not (featurep 'xemacs))
255 (fboundp 'coding-system-list)
256 (fboundp 'sort-coding-systems))
257 (setq mm-mime-mule-charset-alist
258 (apply
259 'nconc
260 (mapcar
261 (lambda (cs)
262 (when (and (or (coding-system-get cs :mime-charset) ; Emacs 22
263 (coding-system-get cs 'mime-charset))
264 (not (eq t (coding-system-get cs 'safe-charsets))))
265 (list (cons (or (coding-system-get cs :mime-charset)
266 (coding-system-get cs 'mime-charset))
267 (delq 'ascii
268 (coding-system-get cs 'safe-charsets))))))
269 (sort-coding-systems (coding-system-list 'base-only))))))
270
271 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
272 "A list of special charsets.
273 Valid elements include:
274 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
275 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
276 )
277
278 (defvar mm-iso-8859-15-compatible
279 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
280 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
281 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
282
283 (defvar mm-iso-8859-x-to-15-table
284 (and (fboundp 'coding-system-p)
285 (mm-coding-system-p 'iso-8859-15)
286 (mapcar
287 (lambda (cs)
288 (if (mm-coding-system-p (car cs))
289 (let ((c (string-to-char
290 (decode-coding-string "\341" (car cs)))))
291 (cons (char-charset c)
292 (cons
293 (- (string-to-char
294 (decode-coding-string "\341" 'iso-8859-15)) c)
295 (string-to-list (decode-coding-string (car (cdr cs))
296 (car cs))))))
297 '(gnus-charset 0)))
298 mm-iso-8859-15-compatible))
299 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
300
301 (defcustom mm-coding-system-priorities
302 (if (boundp 'current-language-environment)
303 (let ((lang (symbol-value 'current-language-environment)))
304 (cond ((string= lang "Japanese")
305 ;; Japanese users may prefer iso-2022-jp to shift-jis.
306 '(iso-2022-jp iso-2022-jp-2 japanese-shift-jis
307 iso-latin-1 utf-8)))))
308 "Preferred coding systems for encoding outgoing messages.
309
310 More than one suitable coding system may be found for some text.
311 By default, the coding system with the highest priority is used
312 to encode outgoing messages (see `sort-coding-systems'). If this
313 variable is set, it overrides the default priority."
314 :version "21.2"
315 :type '(repeat (symbol :tag "Coding system"))
316 :group 'mime)
317
318 ;; ??
319 (defvar mm-use-find-coding-systems-region
320 (fboundp 'find-coding-systems-region)
321 "Use `find-coding-systems-region' to find proper coding systems.
322
323 Setting it to nil is useful on Emacsen supporting Unicode if sending
324 mail with multiple parts is preferred to sending a Unicode one.")
325
326 ;;; Internal variables:
327
328 ;;; Functions:
329
330 (defun mm-mule-charset-to-mime-charset (charset)
331 "Return the MIME charset corresponding to the given Mule CHARSET."
332 (if (and (fboundp 'find-coding-systems-for-charsets)
333 (fboundp 'sort-coding-systems))
334 (let (mime)
335 (dolist (cs (sort-coding-systems
336 (copy-sequence
337 (find-coding-systems-for-charsets (list charset)))))
338 (unless mime
339 (when cs
340 (setq mime (or (coding-system-get cs :mime-charset)
341 (coding-system-get cs 'mime-charset))))))
342 mime)
343 (let ((alist mm-mime-mule-charset-alist)
344 out)
345 (while alist
346 (when (memq charset (cdar alist))
347 (setq out (caar alist)
348 alist nil))
349 (pop alist))
350 out)))
351
352 (defun mm-charset-to-coding-system (charset &optional lbt)
353 "Return coding-system corresponding to CHARSET.
354 CHARSET is a symbol naming a MIME charset.
355 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
356 used as the line break code type of the coding system."
357 (when (stringp charset)
358 (setq charset (intern (downcase charset))))
359 (when lbt
360 (setq charset (intern (format "%s-%s" charset lbt))))
361 (cond
362 ((null charset)
363 charset)
364 ;; Running in a non-MULE environment.
365 ((or (null (mm-get-coding-system-list))
366 (not (fboundp 'coding-system-get)))
367 charset)
368 ;; ascii
369 ((eq charset 'us-ascii)
370 'ascii)
371 ;; Check to see whether we can handle this charset. (This depends
372 ;; on there being some coding system matching each `mime-charset'
373 ;; property defined, as there should be.)
374 ((and (mm-coding-system-p charset)
375 ;;; Doing this would potentially weed out incorrect charsets.
376 ;;; charset
377 ;;; (eq charset (coding-system-get charset 'mime-charset))
378 )
379 charset)
380 ;; Translate invalid charsets.
381 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
382 (and cs (mm-coding-system-p cs) cs)))
383 ;; Last resort: search the coding system list for entries which
384 ;; have the right mime-charset in case the canonical name isn't
385 ;; defined (though it should be).
386 ((let (cs)
387 ;; mm-get-coding-system-list returns a list of cs without lbt.
388 ;; Do we need -lbt?
389 (dolist (c (mm-get-coding-system-list))
390 (if (and (null cs)
391 (eq charset (or (coding-system-get c :mime-charset)
392 (coding-system-get c 'mime-charset))))
393 (setq cs c)))
394 cs))))
395
396 (defsubst mm-replace-chars-in-string (string from to)
397 (mm-subst-char-in-string from to string))
398
399 (eval-and-compile
400 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
401 (boundp 'default-enable-multibyte-characters)
402 default-enable-multibyte-characters
403 (fboundp 'set-buffer-multibyte))
404 "True in Emacs with Mule.")
405
406 (if mm-emacs-mule
407 (defun mm-enable-multibyte ()
408 "Set the multibyte flag of the current buffer.
409 Only do this if the default value of `enable-multibyte-characters' is
410 non-nil. This is a no-op in XEmacs."
411 (set-buffer-multibyte 'to))
412 (defalias 'mm-enable-multibyte 'ignore))
413
414 (if mm-emacs-mule
415 (defun mm-disable-multibyte ()
416 "Unset the multibyte flag of in the current buffer.
417 This is a no-op in XEmacs."
418 (set-buffer-multibyte nil))
419 (defalias 'mm-disable-multibyte 'ignore)))
420
421 (defun mm-preferred-coding-system (charset)
422 ;; A typo in some Emacs versions.
423 (or (get-charset-property charset 'preferred-coding-system)
424 (get-charset-property charset 'prefered-coding-system)))
425
426 ;; Mule charsets shouldn't be used.
427 (defsubst mm-guess-charset ()
428 "Guess Mule charset from the language environment."
429 (or
430 mail-parse-mule-charset ;; cached mule-charset
431 (progn
432 (setq mail-parse-mule-charset
433 (and (boundp 'current-language-environment)
434 (car (last
435 (assq 'charset
436 (assoc current-language-environment
437 language-info-alist))))))
438 (if (or (not mail-parse-mule-charset)
439 (eq mail-parse-mule-charset 'ascii))
440 (setq mail-parse-mule-charset
441 (or (car (last (assq mail-parse-charset
442 mm-mime-mule-charset-alist)))
443 ;; default
444 'latin-iso8859-1)))
445 mail-parse-mule-charset)))
446
447 (defun mm-charset-after (&optional pos)
448 "Return charset of a character in current buffer at position POS.
449 If POS is nil, it defauls to the current point.
450 If POS is out of range, the value is nil.
451 If the charset is `composition', return the actual one."
452 (let ((char (char-after pos)) charset)
453 (if (< (mm-char-int char) 128)
454 (setq charset 'ascii)
455 ;; charset-after is fake in some Emacsen.
456 (setq charset (and (fboundp 'char-charset) (char-charset char)))
457 (if (eq charset 'composition) ; Mule 4
458 (let ((p (or pos (point))))
459 (cadr (find-charset-region p (1+ p))))
460 (if (and charset (not (memq charset '(ascii eight-bit-control
461 eight-bit-graphic))))
462 charset
463 (mm-guess-charset))))))
464
465 (defun mm-mime-charset (charset)
466 "Return the MIME charset corresponding to the given Mule CHARSET."
467 (if (eq charset 'unknown)
468 (error "The message contains non-printable characters, please use attachment"))
469 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
470 ;; This exists in Emacs 20.
471 (or
472 (and (mm-preferred-coding-system charset)
473 (or (coding-system-get
474 (mm-preferred-coding-system charset) :mime-charset)
475 (coding-system-get
476 (mm-preferred-coding-system charset) 'mime-charset)))
477 (and (eq charset 'ascii)
478 'us-ascii)
479 (mm-preferred-coding-system charset)
480 (mm-mule-charset-to-mime-charset charset))
481 ;; This is for XEmacs.
482 (mm-mule-charset-to-mime-charset charset)))
483
484 (defun mm-delete-duplicates (list)
485 "Simple substitute for CL `delete-duplicates', testing with `equal'."
486 (let (result head)
487 (while list
488 (setq head (car list))
489 (setq list (delete head list))
490 (setq result (cons head result)))
491 (nreverse result)))
492
493 ;; Fixme: This is used in places when it should be testing the
494 ;; default multibyteness. See mm-default-multibyte-p.
495 (eval-and-compile
496 (if (and (not (featurep 'xemacs))
497 (boundp 'enable-multibyte-characters))
498 (defun mm-multibyte-p ()
499 "Non-nil if multibyte is enabled in the current buffer."
500 enable-multibyte-characters)
501 (defun mm-multibyte-p () (featurep 'mule))))
502
503 (defun mm-default-multibyte-p ()
504 "Return non-nil if the session is multibyte.
505 This affects whether coding conversion should be attempted generally."
506 (if (featurep 'mule)
507 (if (boundp 'default-enable-multibyte-characters)
508 default-enable-multibyte-characters
509 t)))
510
511 (defun mm-iso-8859-x-to-15-region (&optional b e)
512 (if (fboundp 'char-charset)
513 (let (charset item c inconvertible)
514 (save-restriction
515 (if e (narrow-to-region b e))
516 (goto-char (point-min))
517 (skip-chars-forward "\0-\177")
518 (while (not (eobp))
519 (cond
520 ((not (setq item (assq (char-charset (setq c (char-after)))
521 mm-iso-8859-x-to-15-table)))
522 (forward-char))
523 ((memq c (cdr (cdr item)))
524 (setq inconvertible t)
525 (forward-char))
526 (t
527 (insert-before-markers (prog1 (+ c (car (cdr item)))
528 (delete-char 1)))))
529 (skip-chars-forward "\0-\177")))
530 (not inconvertible))))
531
532 (defun mm-sort-coding-systems-predicate (a b)
533 (let ((priorities
534 (mapcar (lambda (cs)
535 ;; Note: invalid entries are dropped silently
536 (and (coding-system-p cs)
537 (coding-system-base cs)))
538 mm-coding-system-priorities)))
539 (> (length (memq a priorities))
540 (length (memq b priorities)))))
541
542 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
543 "Return the MIME charsets needed to encode the region between B and E.
544 nil means ASCII, a single-element list represents an appropriate MIME
545 charset, and a longer list means no appropriate charset."
546 (let (charsets)
547 ;; The return possibilities of this function are a mess...
548 (or (and (mm-multibyte-p)
549 mm-use-find-coding-systems-region
550 ;; Find the mime-charset of the most preferred coding
551 ;; system that has one.
552 (let ((systems (find-coding-systems-region b e)))
553 (when mm-coding-system-priorities
554 (setq systems
555 (sort systems 'mm-sort-coding-systems-predicate)))
556 (setq systems (delq 'compound-text systems))
557 (unless (equal systems '(undecided))
558 (while systems
559 (let* ((head (pop systems))
560 (cs (or (coding-system-get head :mime-charset)
561 (coding-system-get head 'mime-charset))))
562 ;; The mime-charset (`x-ctext') of
563 ;; `compound-text' is not in the IANA list. We
564 ;; shouldn't normally use anything here with a
565 ;; mime-charset having an `x-' prefix.
566 ;; Fixme: Allow this to be overridden, since
567 ;; there is existing use of x-ctext.
568 ;; Also people apparently need the coding system
569 ;; `iso-2022-jp-3' (which Mule-UCS defines with
570 ;; mime-charset, though it's not valid).
571 (if (and cs
572 (not (string-match "^[Xx]-" (symbol-name cs)))
573 ;; UTF-16 of any variety is invalid for
574 ;; text parts and, unfortunately, has
575 ;; mime-charset defined both in Mule-UCS
576 ;; and versions of Emacs. (The name
577 ;; might be `mule-utf-16...' or
578 ;; `utf-16...'.)
579 (not (string-match "utf-16" (symbol-name cs))))
580 (setq systems nil
581 charsets (list cs))))))
582 charsets))
583 ;; Otherwise we're not multibyte, we're XEmacs, or a single
584 ;; coding system won't cover it.
585 (setq charsets
586 (mm-delete-duplicates
587 (mapcar 'mm-mime-charset
588 (delq 'ascii
589 (mm-find-charset-region b e))))))
590 (if (and (> (length charsets) 1)
591 (memq 'iso-8859-15 charsets)
592 (memq 'iso-8859-15 hack-charsets)
593 (save-excursion (mm-iso-8859-x-to-15-region b e)))
594 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
595 mm-iso-8859-15-compatible))
596 (if (and (memq 'iso-2022-jp-2 charsets)
597 (memq 'iso-2022-jp-2 hack-charsets))
598 (setq charsets (delq 'iso-2022-jp charsets)))
599 charsets))
600
601 (defmacro mm-with-unibyte-buffer (&rest forms)
602 "Create a temporary buffer, and evaluate FORMS there like `progn'.
603 Use unibyte mode for this."
604 `(let (default-enable-multibyte-characters)
605 (with-temp-buffer ,@forms)))
606 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
607 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
608
609 (defmacro mm-with-multibyte-buffer (&rest forms)
610 "Create a temporary buffer, and evaluate FORMS there like `progn'.
611 Use multibyte mode for this."
612 `(let ((default-enable-multibyte-characters t))
613 (with-temp-buffer ,@forms)))
614 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
615 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
616
617 (defmacro mm-with-unibyte-current-buffer (&rest forms)
618 "Evaluate FORMS with current buffer temporarily made unibyte.
619 Also bind `default-enable-multibyte-characters' to nil.
620 Equivalent to `progn' in XEmacs"
621 (let ((multibyte (make-symbol "multibyte"))
622 (buffer (make-symbol "buffer")))
623 `(if mm-emacs-mule
624 (let ((,multibyte enable-multibyte-characters)
625 (,buffer (current-buffer)))
626 (unwind-protect
627 (let (default-enable-multibyte-characters)
628 (set-buffer-multibyte nil)
629 ,@forms)
630 (set-buffer ,buffer)
631 (set-buffer-multibyte ,multibyte)))
632 (let (default-enable-multibyte-characters)
633 ,@forms))))
634 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
635 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
636
637 (defmacro mm-with-unibyte (&rest forms)
638 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
639 `(let (default-enable-multibyte-characters)
640 ,@forms))
641 (put 'mm-with-unibyte 'lisp-indent-function 0)
642 (put 'mm-with-unibyte 'edebug-form-spec '(body))
643
644 (defmacro mm-with-multibyte (&rest forms)
645 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
646 `(let ((default-enable-multibyte-characters t))
647 ,@forms))
648 (put 'mm-with-multibyte 'lisp-indent-function 0)
649 (put 'mm-with-multibyte 'edebug-form-spec '(body))
650
651 (defun mm-find-charset-region (b e)
652 "Return a list of Emacs charsets in the region B to E."
653 (cond
654 ((and (mm-multibyte-p)
655 (fboundp 'find-charset-region))
656 ;; Remove composition since the base charsets have been included.
657 ;; Remove eight-bit-*, treat them as ascii.
658 (let ((css (find-charset-region b e)))
659 (mapcar (lambda (cs) (setq css (delq cs css)))
660 '(composition eight-bit-control eight-bit-graphic
661 control-1))
662 css))
663 (t
664 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
665 (save-excursion
666 (save-restriction
667 (narrow-to-region b e)
668 (goto-char (point-min))
669 (skip-chars-forward "\0-\177")
670 (if (eobp)
671 '(ascii)
672 (let (charset)
673 (setq charset
674 (and (boundp 'current-language-environment)
675 (car (last (assq 'charset
676 (assoc current-language-environment
677 language-info-alist))))))
678 (if (eq charset 'ascii) (setq charset nil))
679 (or charset
680 (setq charset
681 (car (last (assq mail-parse-charset
682 mm-mime-mule-charset-alist)))))
683 (list 'ascii (or charset 'latin-iso8859-1)))))))))
684
685 (if (fboundp 'shell-quote-argument)
686 (defalias 'mm-quote-arg 'shell-quote-argument)
687 (defun mm-quote-arg (arg)
688 "Return a version of ARG that is safe to evaluate in a shell."
689 (let ((pos 0) new-pos accum)
690 ;; *** bug: we don't handle newline characters properly
691 (while (setq new-pos (string-match "[]*[;!'`\"$\\& \t{} |()<>]" arg pos))
692 (push (substring arg pos new-pos) accum)
693 (push "\\" accum)
694 (push (list (aref arg new-pos)) accum)
695 (setq pos (1+ new-pos)))
696 (if (= pos 0)
697 arg
698 (apply 'concat (nconc (nreverse accum) (list (substring arg pos))))))))
699
700 (defun mm-auto-mode-alist ()
701 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
702 (let ((alist auto-mode-alist)
703 out)
704 (while alist
705 (when (listp (cdar alist))
706 (push (car alist) out))
707 (pop alist))
708 (nreverse out)))
709
710 (defvar mm-inhibit-file-name-handlers
711 '(jka-compr-handler image-file-handler)
712 "A list of handlers doing (un)compression (etc) thingies.")
713
714 (defun mm-insert-file-contents (filename &optional visit beg end replace
715 inhibit)
716 "Like `insert-file-contents', but only reads in the file.
717 A buffer may be modified in several ways after reading into the buffer due
718 to advanced Emacs features, such as file-name-handlers, format decoding,
719 `find-file-hooks', etc.
720 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
721 This function ensures that none of these modifications will take place."
722 (let ((format-alist nil)
723 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
724 (default-major-mode 'fundamental-mode)
725 (enable-local-variables nil)
726 (after-insert-file-functions nil)
727 (enable-local-eval nil)
728 (find-file-hooks nil)
729 (inhibit-file-name-operation (if inhibit
730 'insert-file-contents
731 inhibit-file-name-operation))
732 (inhibit-file-name-handlers
733 (if inhibit
734 (append mm-inhibit-file-name-handlers
735 inhibit-file-name-handlers)
736 inhibit-file-name-handlers)))
737 (insert-file-contents filename visit beg end replace)))
738
739 (defun mm-append-to-file (start end filename &optional codesys inhibit)
740 "Append the contents of the region to the end of file FILENAME.
741 When called from a function, expects three arguments,
742 START, END and FILENAME. START and END are buffer positions
743 saying what text to write.
744 Optional fourth argument specifies the coding system to use when
745 encoding the file.
746 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
747 (let ((coding-system-for-write
748 (or codesys mm-text-coding-system-for-write
749 mm-text-coding-system))
750 (inhibit-file-name-operation (if inhibit
751 'append-to-file
752 inhibit-file-name-operation))
753 (inhibit-file-name-handlers
754 (if inhibit
755 (append mm-inhibit-file-name-handlers
756 inhibit-file-name-handlers)
757 inhibit-file-name-handlers)))
758 (write-region start end filename t 'no-message)
759 (message "Appended to %s" filename)))
760
761 (defun mm-write-region (start end filename &optional append visit lockname
762 coding-system inhibit)
763
764 "Like `write-region'.
765 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
766 (let ((coding-system-for-write
767 (or coding-system mm-text-coding-system-for-write
768 mm-text-coding-system))
769 (inhibit-file-name-operation (if inhibit
770 'write-region
771 inhibit-file-name-operation))
772 (inhibit-file-name-handlers
773 (if inhibit
774 (append mm-inhibit-file-name-handlers
775 inhibit-file-name-handlers)
776 inhibit-file-name-handlers)))
777 (write-region start end filename append visit lockname)))
778
779 (defun mm-image-load-path (&optional package)
780 (let (dir result)
781 (dolist (path load-path (nreverse result))
782 (when (and path
783 (file-directory-p
784 (setq dir (concat (file-name-directory
785 (directory-file-name path))
786 "etc/" (or package "gnus/")))))
787 (push dir result))
788 (push path result))))
789
790 ;; Fixme: This doesn't look useful where it's used.
791 (if (fboundp 'detect-coding-region)
792 (defun mm-detect-coding-region (start end)
793 "Like `detect-coding-region' except returning the best one."
794 (let ((coding-systems
795 (detect-coding-region (point) (point-max))))
796 (or (car-safe coding-systems)
797 coding-systems)))
798 (defun mm-detect-coding-region (start end)
799 (let ((point (point)))
800 (goto-char start)
801 (skip-chars-forward "\0-\177" end)
802 (prog1
803 (if (eq (point) end) 'ascii (mm-guess-charset))
804 (goto-char point)))))
805
806 (if (fboundp 'coding-system-get)
807 (defun mm-detect-mime-charset-region (start end)
808 "Detect MIME charset of the text in the region between START and END."
809 (let ((cs (mm-detect-coding-region start end)))
810 (coding-system-get cs 'mime-charset)))
811 (defun mm-detect-mime-charset-region (start end)
812 "Detect MIME charset of the text in the region between START and END."
813 (let ((cs (mm-detect-coding-region start end)))
814 cs)))
815
816
817 (provide 'mm-util)
818
819 ;;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
820 ;;; mm-util.el ends here