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.
37 (let ((nfunc (intern (format "mm-%s" (car elem)))))
38 (if (fboundp (car elem))
39 (defalias nfunc (car elem))
40 (defalias nfunc (cdr elem)))))
41 '((decode-coding-string . (lambda (s a) s))
42 (encode-coding-string . (lambda (s a) s))
43 (encode-coding-region . ignore)
44 (coding-system-list . ignore)
45 (decode-coding-region . ignore)
47 (coding-system-equal . equal)
48 (annotationp . ignore)
49 (set-buffer-file-coding-system . ignore)
56 (mapcar (lambda (e) (list (symbol-name (car e))))
57 mm-mime-mule-charset-alist)
60 . (lambda (from to string &optional inplace) ;; stolen (and renamed) from nnheader.el
61 "Replace characters in STRING from FROM to TO.
62 Unless optional argument INPLACE is non-nil, return a new string."
63 (let ((string (if inplace string (copy-sequence string)))
66 ;; Replace all occurrences of FROM with TO.
68 (when (= (aref string idx) from)
73 . (lambda (string regexp rep &optional literal)
74 "See `replace-regexp-in-string', only the order of args differs."
75 (replace-regexp-in-string regexp rep string nil literal)))
76 (string-as-unibyte . identity)
77 (string-make-unibyte . identity)
78 (string-as-multibyte . identity)
79 (multibyte-string-p . ignore)
80 ;; It is not a MIME function, but some MIME functions use it.
81 (make-temp-file . (lambda (prefix &optional dir-flag)
82 (let ((file (expand-file-name
83 (make-temp-name prefix)
84 (if (fboundp 'temp-directory)
86 temporary-file-directory))))
88 (make-directory file))
90 (insert-byte . insert-char)
91 (multibyte-char-to-unibyte . identity))))
94 (defalias 'mm-char-or-char-int-p
96 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
97 ((fboundp 'char-valid-p) 'char-valid-p)
100 ;; Fixme: This seems always to be used to read a MIME charset, so it
101 ;; should be re-named and fixed (in Emacs) to offer completion only on
102 ;; proper charset names (base coding systems which have a
103 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
105 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
106 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
107 ;; Actually, there should be an `mm-coding-system-mime-charset'.
109 (defalias 'mm-read-coding-system
111 ((fboundp 'read-coding-system)
112 (if (and (featurep 'xemacs)
113 (<= (string-to-number emacs-version) 21.1))
114 (lambda (prompt &optional default-coding-system)
115 (read-coding-system prompt))
116 'read-coding-system))
117 (t (lambda (prompt &optional default-coding-system)
118 "Prompt the user for a coding system."
120 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
121 mm-mime-mule-charset-alist)))))))
123 (defvar mm-coding-system-list nil)
124 (defun mm-get-coding-system-list ()
125 "Get the coding system list."
126 (or mm-coding-system-list
127 (setq mm-coding-system-list (mm-coding-system-list))))
129 (defun mm-coding-system-p (cs)
130 "Return non-nil if CS is a symbol naming a coding system.
131 In XEmacs, also return non-nil if CS is a coding system object.
132 If CS is available, return CS itself in Emacs, and return a coding
133 system object in XEmacs."
134 (if (fboundp 'find-coding-system)
135 (find-coding-system cs)
136 (if (fboundp 'coding-system-p)
137 (when (coding-system-p cs)
139 ;; Is this branch ever actually useful?
140 (car (memq cs (mm-get-coding-system-list))))))
142 (defvar mm-charset-synonym-alist
144 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
145 ,@(unless (mm-coding-system-p 'x-ctext)
146 '((x-ctext . ctext)))
147 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_!
148 ,@(unless (mm-coding-system-p 'iso-8859-15)
149 '((iso-8859-15 . iso-8859-1)))
150 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
151 ,@(unless (mm-coding-system-p 'big5-hkscs)
152 '((big5-hkscs . big5)))
153 ;; Windows-1252 is actually a superset of Latin-1. See also
154 ;; `gnus-article-dumbquotes-map'.
155 ,@(unless (mm-coding-system-p 'windows-1252)
156 (if (mm-coding-system-p 'cp1252)
157 '((windows-1252 . cp1252))
158 '((windows-1252 . iso-8859-1))))
159 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
160 ;; Outlook users in Czech republic. Use this to allow reading of their
161 ;; e-mails. cp1250 should be defined by M-x codepage-setup.
162 ,@(if (and (not (mm-coding-system-p 'windows-1250))
163 (mm-coding-system-p 'cp1250))
164 '((windows-1250 . cp1250)))
165 ;; A Microsoft misunderstanding.
166 ,@(if (and (not (mm-coding-system-p 'unicode))
167 (mm-coding-system-p 'utf-16-le))
168 '((unicode . utf-16-le)))
169 ;; A Microsoft misunderstanding.
170 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
171 (if (mm-coding-system-p 'cp949)
172 '((ks_c_5601-1987 . cp949))
173 '((ks_c_5601-1987 . euc-kr))))
175 "A mapping from invalid charset names to the real charset names.")
177 (defvar mm-binary-coding-system
179 ((mm-coding-system-p 'binary) 'binary)
180 ((mm-coding-system-p 'no-conversion) 'no-conversion)
182 "100% binary coding system.")
184 (defvar mm-text-coding-system
185 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
186 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
187 (and (mm-coding-system-p 'raw-text) 'raw-text))
188 mm-binary-coding-system)
189 "Text-safe coding system (For removing ^M).")
191 (defvar mm-text-coding-system-for-write nil
192 "Text coding system for write.")
194 (defvar mm-auto-save-coding-system
196 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
197 (if (memq system-type '(windows-nt ms-dos ms-windows))
198 (if (mm-coding-system-p 'utf-8-emacs-dos)
199 'utf-8-emacs-dos mm-binary-coding-system)
201 ((mm-coding-system-p 'emacs-mule)
202 (if (memq system-type '(windows-nt ms-dos ms-windows))
203 (if (mm-coding-system-p 'emacs-mule-dos)
204 'emacs-mule-dos mm-binary-coding-system)
206 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
207 (t mm-binary-coding-system))
208 "Coding system of auto save file.")
210 (defvar mm-universal-coding-system mm-auto-save-coding-system
211 "The universal coding system.")
213 ;; Fixme: some of the cars here aren't valid MIME charsets. That
214 ;; should only matter with XEmacs, though.
215 (defvar mm-mime-mule-charset-alist
217 (iso-8859-1 latin-iso8859-1)
218 (iso-8859-2 latin-iso8859-2)
219 (iso-8859-3 latin-iso8859-3)
220 (iso-8859-4 latin-iso8859-4)
221 (iso-8859-5 cyrillic-iso8859-5)
222 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
223 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
224 ;; charset is koi8-r, not iso-8859-5.
225 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
226 (iso-8859-6 arabic-iso8859-6)
227 (iso-8859-7 greek-iso8859-7)
228 (iso-8859-8 hebrew-iso8859-8)
229 (iso-8859-9 latin-iso8859-9)
230 (iso-8859-14 latin-iso8859-14)
231 (iso-8859-15 latin-iso8859-15)
232 (viscii vietnamese-viscii-lower)
233 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
234 (euc-kr korean-ksc5601)
235 (gb2312 chinese-gb2312)
236 (big5 chinese-big5-1 chinese-big5-2)
238 (thai-tis620 thai-tis620)
239 (windows-1251 cyrillic-iso8859-5)
240 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
241 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
242 latin-jisx0201 japanese-jisx0208-1978
243 chinese-gb2312 japanese-jisx0208
244 korean-ksc5601 japanese-jisx0212)
245 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
246 latin-jisx0201 japanese-jisx0208-1978
247 chinese-gb2312 japanese-jisx0208
248 korean-ksc5601 japanese-jisx0212
249 chinese-cns11643-1 chinese-cns11643-2)
250 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
251 cyrillic-iso8859-5 greek-iso8859-7
252 latin-jisx0201 japanese-jisx0208-1978
253 chinese-gb2312 japanese-jisx0208
254 korean-ksc5601 japanese-jisx0212
255 chinese-cns11643-1 chinese-cns11643-2
256 chinese-cns11643-3 chinese-cns11643-4
257 chinese-cns11643-5 chinese-cns11643-6
259 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
260 japanese-jisx0213-1 japanese-jisx0213-2)
261 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
262 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
263 (charsetp 'unicode-a)
264 (not (mm-coding-system-p 'mule-utf-8)))
265 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
266 ;; If we have utf-8 we're in Mule 5+.
269 (coding-system-get 'mule-utf-8 'safe-charsets)))))
270 "Alist of MIME-charset/MULE-charsets.")
272 (defun mm-enrich-utf-8-by-mule-ucs ()
273 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
274 This function will run when the `un-define' module is loaded under
275 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
276 with Mule charsets. It is completely useless for Emacs."
277 (unless (cdr (delete '(mm-enrich-utf-8-by-mule-ucs)
278 (assoc "un-define" after-load-alist)))
279 (setq after-load-alist
280 (delete '("un-define") after-load-alist)))
281 (when (boundp 'unicode-basic-translation-charset-order-list)
287 'unicode-basic-translation-charset-order-list))))
288 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
291 (setq mm-mime-mule-charset-alist
292 (nconc mm-mime-mule-charset-alist
293 (list (cons 'utf-8 val))))))
296 ;; Correct by construction, but should be unnecessary for Emacs:
297 (if (featurep 'xemacs)
298 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
299 (when (and (fboundp 'coding-system-list)
300 (fboundp 'sort-coding-systems))
301 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
305 mime (or (coding-system-get cs :mime-charset) ; Emacs 22
306 (coding-system-get cs 'mime-charset)))
308 (not (eq t (setq mule
309 (coding-system-get cs 'safe-charsets))))
310 (not (assq mime alist)))
311 (push (cons mime (delq 'ascii mule)) alist)))
312 (setq mm-mime-mule-charset-alist (nreverse alist)))))
314 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
315 "A list of special charsets.
316 Valid elements include:
317 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
318 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
321 (defvar mm-iso-8859-15-compatible
322 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
323 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
324 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
326 (defvar mm-iso-8859-x-to-15-table
327 (and (fboundp 'coding-system-p)
328 (mm-coding-system-p 'iso-8859-15)
331 (if (mm-coding-system-p (car cs))
332 (let ((c (string-to-char
333 (decode-coding-string "\341" (car cs)))))
334 (cons (char-charset c)
337 (decode-coding-string "\341" 'iso-8859-15)) c)
338 (string-to-list (decode-coding-string (car (cdr cs))
341 mm-iso-8859-15-compatible))
342 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
344 (defcustom mm-coding-system-priorities
345 (if (boundp 'current-language-environment)
346 (let ((lang (symbol-value 'current-language-environment)))
347 (cond ((string= lang "Japanese")
348 ;; Japanese users prefer iso-2022-jp to euc-japan or
349 ;; shift_jis, however iso-8859-1 should be used when
350 ;; there are only ASCII text and Latin-1 characters.
351 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
352 "Preferred coding systems for encoding outgoing messages.
354 More than one suitable coding system may be found for some text.
355 By default, the coding system with the highest priority is used
356 to encode outgoing messages (see `sort-coding-systems'). If this
357 variable is set, it overrides the default priority."
359 :type '(repeat (symbol :tag "Coding system"))
363 (defvar mm-use-find-coding-systems-region
364 (fboundp 'find-coding-systems-region)
365 "Use `find-coding-systems-region' to find proper coding systems.
367 Setting it to nil is useful on Emacsen supporting Unicode if sending
368 mail with multiple parts is preferred to sending a Unicode one.")
370 ;;; Internal variables:
374 (defun mm-mule-charset-to-mime-charset (charset)
375 "Return the MIME charset corresponding to the given Mule CHARSET."
376 (if (and (fboundp 'find-coding-systems-for-charsets)
377 (fboundp 'sort-coding-systems))
378 (let ((css (sort (sort-coding-systems
379 (find-coding-systems-for-charsets (list charset)))
380 'mm-sort-coding-systems-predicate))
382 (while (and (not mime)
384 (when (setq cs (pop css))
385 (setq mime (or (coding-system-get cs :mime-charset)
386 (coding-system-get cs 'mime-charset)))))
388 (let ((alist (mapcar (lambda (cs)
389 (assq cs mm-mime-mule-charset-alist))
390 (sort (mapcar 'car mm-mime-mule-charset-alist)
391 'mm-sort-coding-systems-predicate)))
394 (when (memq charset (cdar alist))
395 (setq out (caar alist)
400 (defun mm-charset-to-coding-system (charset &optional lbt)
401 "Return coding-system corresponding to CHARSET.
402 CHARSET is a symbol naming a MIME charset.
403 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
404 used as the line break code type of the coding system."
405 (when (stringp charset)
406 (setq charset (intern (downcase charset))))
408 (setq charset (intern (format "%s-%s" charset lbt))))
412 ;; Running in a non-MULE environment.
413 ((or (null (mm-get-coding-system-list))
414 (not (fboundp 'coding-system-get)))
417 ((eq charset 'us-ascii)
419 ;; Check to see whether we can handle this charset. (This depends
420 ;; on there being some coding system matching each `mime-charset'
421 ;; property defined, as there should be.)
422 ((and (mm-coding-system-p charset)
423 ;;; Doing this would potentially weed out incorrect charsets.
425 ;;; (eq charset (coding-system-get charset 'mime-charset))
428 ;; Translate invalid charsets.
429 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
430 (and cs (mm-coding-system-p cs) cs)))
431 ;; Last resort: search the coding system list for entries which
432 ;; have the right mime-charset in case the canonical name isn't
433 ;; defined (though it should be).
435 ;; mm-get-coding-system-list returns a list of cs without lbt.
437 (dolist (c (mm-get-coding-system-list))
439 (eq charset (or (coding-system-get c :mime-charset)
440 (coding-system-get c 'mime-charset))))
445 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
446 (boundp 'default-enable-multibyte-characters)
447 default-enable-multibyte-characters
448 (fboundp 'set-buffer-multibyte))
449 "True in Emacs with Mule.")
452 (defun mm-enable-multibyte ()
453 "Set the multibyte flag of the current buffer.
454 Only do this if the default value of `enable-multibyte-characters' is
455 non-nil. This is a no-op in XEmacs."
456 (set-buffer-multibyte 'to))
457 (defalias 'mm-enable-multibyte 'ignore))
460 (defun mm-disable-multibyte ()
461 "Unset the multibyte flag of in the current buffer.
462 This is a no-op in XEmacs."
463 (set-buffer-multibyte nil))
464 (defalias 'mm-disable-multibyte 'ignore)))
466 (defun mm-preferred-coding-system (charset)
467 ;; A typo in some Emacs versions.
468 (or (get-charset-property charset 'preferred-coding-system)
469 (get-charset-property charset 'prefered-coding-system)))
471 ;; Mule charsets shouldn't be used.
472 (defsubst mm-guess-charset ()
473 "Guess Mule charset from the language environment."
475 mail-parse-mule-charset ;; cached mule-charset
477 (setq mail-parse-mule-charset
478 (and (boundp 'current-language-environment)
481 (assoc current-language-environment
482 language-info-alist))))))
483 (if (or (not mail-parse-mule-charset)
484 (eq mail-parse-mule-charset 'ascii))
485 (setq mail-parse-mule-charset
486 (or (car (last (assq mail-parse-charset
487 mm-mime-mule-charset-alist)))
490 mail-parse-mule-charset)))
492 (defun mm-charset-after (&optional pos)
493 "Return charset of a character in current buffer at position POS.
494 If POS is nil, it defauls to the current point.
495 If POS is out of range, the value is nil.
496 If the charset is `composition', return the actual one."
497 (let ((char (char-after pos)) charset)
498 (if (< (mm-char-int char) 128)
499 (setq charset 'ascii)
500 ;; charset-after is fake in some Emacsen.
501 (setq charset (and (fboundp 'char-charset) (char-charset char)))
502 (if (eq charset 'composition) ; Mule 4
503 (let ((p (or pos (point))))
504 (cadr (find-charset-region p (1+ p))))
505 (if (and charset (not (memq charset '(ascii eight-bit-control
506 eight-bit-graphic))))
508 (mm-guess-charset))))))
510 (defun mm-mime-charset (charset)
511 "Return the MIME charset corresponding to the given Mule CHARSET."
512 (if (eq charset 'unknown)
513 (error "The message contains non-printable characters, please use attachment"))
514 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
515 ;; This exists in Emacs 20.
517 (and (mm-preferred-coding-system charset)
518 (or (coding-system-get
519 (mm-preferred-coding-system charset) :mime-charset)
521 (mm-preferred-coding-system charset) 'mime-charset)))
522 (and (eq charset 'ascii)
524 (mm-preferred-coding-system charset)
525 (mm-mule-charset-to-mime-charset charset))
526 ;; This is for XEmacs.
527 (mm-mule-charset-to-mime-charset charset)))
529 (defun mm-delete-duplicates (list)
530 "Simple substitute for CL `delete-duplicates', testing with `equal'."
533 (setq head (car list))
534 (setq list (delete head list))
535 (setq result (cons head result)))
538 ;; Fixme: This is used in places when it should be testing the
539 ;; default multibyteness. See mm-default-multibyte-p.
541 (if (and (not (featurep 'xemacs))
542 (boundp 'enable-multibyte-characters))
543 (defun mm-multibyte-p ()
544 "Non-nil if multibyte is enabled in the current buffer."
545 enable-multibyte-characters)
546 (defun mm-multibyte-p () (featurep 'mule))))
548 (defun mm-default-multibyte-p ()
549 "Return non-nil if the session is multibyte.
550 This affects whether coding conversion should be attempted generally."
552 (if (boundp 'default-enable-multibyte-characters)
553 default-enable-multibyte-characters
556 (defun mm-iso-8859-x-to-15-region (&optional b e)
557 (if (fboundp 'char-charset)
558 (let (charset item c inconvertible)
560 (if e (narrow-to-region b e))
561 (goto-char (point-min))
562 (skip-chars-forward "\0-\177")
565 ((not (setq item (assq (char-charset (setq c (char-after)))
566 mm-iso-8859-x-to-15-table)))
568 ((memq c (cdr (cdr item)))
569 (setq inconvertible t)
572 (insert-before-markers (prog1 (+ c (car (cdr item)))
574 (skip-chars-forward "\0-\177")))
575 (not inconvertible))))
577 (defun mm-sort-coding-systems-predicate (a b)
580 ;; Note: invalid entries are dropped silently
581 (and (setq cs (mm-coding-system-p cs))
582 (coding-system-base cs)))
583 mm-coding-system-priorities)))
584 (and (setq a (mm-coding-system-p a))
585 (if (setq b (mm-coding-system-p b))
586 (> (length (memq (coding-system-base a) priorities))
587 (length (memq (coding-system-base b) priorities)))
591 (autoload 'latin-unity-massage-name "latin-unity")
592 (autoload 'latin-unity-maybe-remap "latin-unity")
593 (autoload 'latin-unity-representations-feasible-region "latin-unity")
594 (autoload 'latin-unity-representations-present-region "latin-unity")
595 (defvar latin-unity-coding-systems)
596 (defvar latin-unity-ucs-list))
598 (defun mm-xemacs-find-mime-charset-1 (begin end)
599 "Determine which MIME charset to use to send region as message.
600 This uses the XEmacs-specific latin-unity package to better handle the
601 case where identical characters from diverse ISO-8859-? character sets
602 can be encoded using a single one of the corresponding coding systems.
604 It treats `mm-coding-system-priorities' as the list of preferred
605 coding systems; a useful example setting for this list in Western
606 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
607 to the very standard Latin 1 coding system, and only move to coding
608 systems that are less supported as is necessary to encode the
609 characters that exist in the buffer.
611 Latin Unity doesn't know about those non-ASCII Roman characters that
612 are available in various East Asian character sets. As such, its
613 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
614 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
615 But this is very much a corner case, so don't worry about it."
616 (let ((systems mm-coding-system-priorities) csets psets curset)
618 ;; Load the Latin Unity library, if available.
619 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
620 (require 'latin-unity))
622 ;; Now, can we use it?
623 (if (featurep 'latin-unity)
625 (setq csets (latin-unity-representations-feasible-region begin end)
626 psets (latin-unity-representations-present-region begin end))
630 ;; Pass back the first coding system in the preferred list
631 ;; that can encode the whole region.
632 (dolist (curset systems)
633 (setq curset (latin-unity-massage-name 'buffer-default curset))
635 ;; If the coding system is a universal coding system, then
636 ;; it can certainly encode all the characters in the region.
637 (if (memq curset latin-unity-ucs-list)
638 (throw 'done (list curset)))
640 ;; If a coding system isn't universal, and isn't in
641 ;; the list that latin unity knows about, we can't
642 ;; decide whether to use it here. Leave that until later
643 ;; in `mm-find-mime-charset-region' function, whence we
645 (unless (memq curset latin-unity-coding-systems)
648 ;; Right, we know about this coding system, and it may
649 ;; conceivably be able to encode all the characters in
651 (if (latin-unity-maybe-remap begin end curset csets psets t)
652 (throw 'done (list curset))))
654 ;; Can't encode using anything from the
655 ;; `mm-coding-system-priorities' list.
656 ;; Leave `mm-find-mime-charset' to do most of the work.
659 ;; Right, latin unity isn't available; let `mm-find-charset-region'
660 ;; take its default action, which equally applies to GNU Emacs.
663 (defmacro mm-xemacs-find-mime-charset (begin end)
664 (when (featurep 'xemacs)
665 `(mm-xemacs-find-mime-charset-1 ,begin ,end)))
667 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
668 "Return the MIME charsets needed to encode the region between B and E.
669 nil means ASCII, a single-element list represents an appropriate MIME
670 charset, and a longer list means no appropriate charset."
672 ;; The return possibilities of this function are a mess...
673 (or (and (mm-multibyte-p)
674 mm-use-find-coding-systems-region
675 ;; Find the mime-charset of the most preferred coding
676 ;; system that has one.
677 (let ((systems (find-coding-systems-region b e)))
678 (when mm-coding-system-priorities
680 (sort systems 'mm-sort-coding-systems-predicate)))
681 (setq systems (delq 'compound-text systems))
682 (unless (equal systems '(undecided))
684 (let* ((head (pop systems))
685 (cs (or (coding-system-get head :mime-charset)
686 (coding-system-get head 'mime-charset))))
687 ;; The mime-charset (`x-ctext') of
688 ;; `compound-text' is not in the IANA list. We
689 ;; shouldn't normally use anything here with a
690 ;; mime-charset having an `x-' prefix.
691 ;; Fixme: Allow this to be overridden, since
692 ;; there is existing use of x-ctext.
693 ;; Also people apparently need the coding system
694 ;; `iso-2022-jp-3' (which Mule-UCS defines with
695 ;; mime-charset, though it's not valid).
697 (not (string-match "^[Xx]-" (symbol-name cs)))
698 ;; UTF-16 of any variety is invalid for
699 ;; text parts and, unfortunately, has
700 ;; mime-charset defined both in Mule-UCS
701 ;; and versions of Emacs. (The name
702 ;; might be `mule-utf-16...' or
704 (not (string-match "utf-16" (symbol-name cs))))
706 charsets (list cs))))))
708 ;; If we're XEmacs, and some coding system is appropriate,
709 ;; mm-xemacs-find-mime-charset will return an appropriate list.
710 ;; Otherwise, we'll get nil, and the next setq will get invoked.
711 (setq charsets (mm-xemacs-find-mime-charset b e))
713 ;; We're not multibyte, or a single coding system won't cover it.
715 (mm-delete-duplicates
716 (mapcar 'mm-mime-charset
718 (mm-find-charset-region b e))))))
719 (if (and (> (length charsets) 1)
720 (memq 'iso-8859-15 charsets)
721 (memq 'iso-8859-15 hack-charsets)
722 (save-excursion (mm-iso-8859-x-to-15-region b e)))
723 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
724 mm-iso-8859-15-compatible))
725 (if (and (memq 'iso-2022-jp-2 charsets)
726 (memq 'iso-2022-jp-2 hack-charsets))
727 (setq charsets (delq 'iso-2022-jp charsets)))
730 (defmacro mm-with-unibyte-buffer (&rest forms)
731 "Create a temporary buffer, and evaluate FORMS there like `progn'.
732 Use unibyte mode for this."
733 `(let (default-enable-multibyte-characters)
734 (with-temp-buffer ,@forms)))
735 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
736 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
738 (defmacro mm-with-multibyte-buffer (&rest forms)
739 "Create a temporary buffer, and evaluate FORMS there like `progn'.
740 Use multibyte mode for this."
741 `(let ((default-enable-multibyte-characters t))
742 (with-temp-buffer ,@forms)))
743 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
744 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
746 (defmacro mm-with-unibyte-current-buffer (&rest forms)
747 "Evaluate FORMS with current buffer temporarily made unibyte.
748 Also bind `default-enable-multibyte-characters' to nil.
749 Equivalent to `progn' in XEmacs"
750 (let ((multibyte (make-symbol "multibyte"))
751 (buffer (make-symbol "buffer")))
753 (let ((,multibyte enable-multibyte-characters)
754 (,buffer (current-buffer)))
756 (let (default-enable-multibyte-characters)
757 (set-buffer-multibyte nil)
760 (set-buffer-multibyte ,multibyte)))
761 (let (default-enable-multibyte-characters)
763 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
764 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
766 (defmacro mm-with-unibyte (&rest forms)
767 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
768 `(let (default-enable-multibyte-characters)
770 (put 'mm-with-unibyte 'lisp-indent-function 0)
771 (put 'mm-with-unibyte 'edebug-form-spec '(body))
773 (defmacro mm-with-multibyte (&rest forms)
774 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
775 `(let ((default-enable-multibyte-characters t))
777 (put 'mm-with-multibyte 'lisp-indent-function 0)
778 (put 'mm-with-multibyte 'edebug-form-spec '(body))
780 (defun mm-find-charset-region (b e)
781 "Return a list of Emacs charsets in the region B to E."
783 ((and (mm-multibyte-p)
784 (fboundp 'find-charset-region))
785 ;; Remove composition since the base charsets have been included.
786 ;; Remove eight-bit-*, treat them as ascii.
787 (let ((css (find-charset-region b e)))
788 (mapcar (lambda (cs) (setq css (delq cs css)))
789 '(composition eight-bit-control eight-bit-graphic
793 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
796 (narrow-to-region b e)
797 (goto-char (point-min))
798 (skip-chars-forward "\0-\177")
803 (and (boundp 'current-language-environment)
804 (car (last (assq 'charset
805 (assoc current-language-environment
806 language-info-alist))))))
807 (if (eq charset 'ascii) (setq charset nil))
810 (car (last (assq mail-parse-charset
811 mm-mime-mule-charset-alist)))))
812 (list 'ascii (or charset 'latin-iso8859-1)))))))))
814 (defun mm-auto-mode-alist ()
815 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
816 (let ((alist auto-mode-alist)
819 (when (listp (cdar alist))
820 (push (car alist) out))
824 (defvar mm-inhibit-file-name-handlers
825 '(jka-compr-handler image-file-handler)
826 "A list of handlers doing (un)compression (etc) thingies.")
828 (defun mm-insert-file-contents (filename &optional visit beg end replace
830 "Like `insert-file-contents', but only reads in the file.
831 A buffer may be modified in several ways after reading into the buffer due
832 to advanced Emacs features, such as file-name-handlers, format decoding,
833 `find-file-hooks', etc.
834 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
835 This function ensures that none of these modifications will take place."
836 (let ((format-alist nil)
837 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
838 (default-major-mode 'fundamental-mode)
839 (enable-local-variables nil)
840 (after-insert-file-functions nil)
841 (enable-local-eval nil)
842 (find-file-hooks nil)
843 (inhibit-file-name-operation (if inhibit
844 'insert-file-contents
845 inhibit-file-name-operation))
846 (inhibit-file-name-handlers
848 (append mm-inhibit-file-name-handlers
849 inhibit-file-name-handlers)
850 inhibit-file-name-handlers)))
851 (insert-file-contents filename visit beg end replace)))
853 (defun mm-append-to-file (start end filename &optional codesys inhibit)
854 "Append the contents of the region to the end of file FILENAME.
855 When called from a function, expects three arguments,
856 START, END and FILENAME. START and END are buffer positions
857 saying what text to write.
858 Optional fourth argument specifies the coding system to use when
860 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
861 (let ((coding-system-for-write
862 (or codesys mm-text-coding-system-for-write
863 mm-text-coding-system))
864 (inhibit-file-name-operation (if inhibit
866 inhibit-file-name-operation))
867 (inhibit-file-name-handlers
869 (append mm-inhibit-file-name-handlers
870 inhibit-file-name-handlers)
871 inhibit-file-name-handlers)))
872 (write-region start end filename t 'no-message)
873 (message "Appended to %s" filename)))
875 (defun mm-write-region (start end filename &optional append visit lockname
876 coding-system inhibit)
878 "Like `write-region'.
879 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
880 (let ((coding-system-for-write
881 (or coding-system mm-text-coding-system-for-write
882 mm-text-coding-system))
883 (inhibit-file-name-operation (if inhibit
885 inhibit-file-name-operation))
886 (inhibit-file-name-handlers
888 (append mm-inhibit-file-name-handlers
889 inhibit-file-name-handlers)
890 inhibit-file-name-handlers)))
891 (write-region start end filename append visit lockname)))
893 (defun mm-image-load-path (&optional package)
895 (dolist (path load-path (nreverse result))
898 (setq dir (concat (file-name-directory
899 (directory-file-name path))
900 "etc/" (or package "gnus/")))))
902 (push path result))))
904 ;; Fixme: This doesn't look useful where it's used.
905 (if (fboundp 'detect-coding-region)
906 (defun mm-detect-coding-region (start end)
907 "Like `detect-coding-region' except returning the best one."
908 (let ((coding-systems
909 (detect-coding-region (point) (point-max))))
910 (or (car-safe coding-systems)
912 (defun mm-detect-coding-region (start end)
913 (let ((point (point)))
915 (skip-chars-forward "\0-\177" end)
917 (if (eq (point) end) 'ascii (mm-guess-charset))
918 (goto-char point)))))
920 (if (fboundp 'coding-system-get)
921 (defun mm-detect-mime-charset-region (start end)
922 "Detect MIME charset of the text in the region between START and END."
923 (let ((cs (mm-detect-coding-region start end)))
924 (coding-system-get cs 'mime-charset)))
925 (defun mm-detect-mime-charset-region (start end)
926 "Detect MIME charset of the text in the region between START and END."
927 (let ((cs (mm-detect-coding-region start end)))
933 ;;; mm-util.el ends here