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 ,@(if (and (not (mm-coding-system-p 'unicode))
164 (mm-coding-system-p 'utf-16-le))
165 '((unicode . utf-16-le)))
166 ;; A Microsoft misunderstanding.
167 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
168 (if (mm-coding-system-p 'cp949)
169 '((ks_c_5601-1987 . cp949))
170 '((ks_c_5601-1987 . euc-kr))))
172 "A mapping from invalid charset names to the real charset names.")
174 (defvar mm-binary-coding-system
176 ((mm-coding-system-p 'binary) 'binary)
177 ((mm-coding-system-p 'no-conversion) 'no-conversion)
179 "100% binary coding system.")
181 (defvar mm-text-coding-system
182 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
183 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
184 (and (mm-coding-system-p 'raw-text) 'raw-text))
185 mm-binary-coding-system)
186 "Text-safe coding system (For removing ^M).")
188 (defvar mm-text-coding-system-for-write nil
189 "Text coding system for write.")
191 (defvar mm-auto-save-coding-system
193 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
194 (if (memq system-type '(windows-nt ms-dos ms-windows))
195 (if (mm-coding-system-p 'utf-8-emacs-dos)
196 'utf-8-emacs-dos mm-binary-coding-system)
198 ((mm-coding-system-p 'emacs-mule)
199 (if (memq system-type '(windows-nt ms-dos ms-windows))
200 (if (mm-coding-system-p 'emacs-mule-dos)
201 'emacs-mule-dos mm-binary-coding-system)
203 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
204 (t mm-binary-coding-system))
205 "Coding system of auto save file.")
207 (defvar mm-universal-coding-system mm-auto-save-coding-system
208 "The universal coding system.")
210 ;; Fixme: some of the cars here aren't valid MIME charsets. That
211 ;; should only matter with XEmacs, though.
212 (defvar mm-mime-mule-charset-alist
214 (iso-8859-1 latin-iso8859-1)
215 (iso-8859-2 latin-iso8859-2)
216 (iso-8859-3 latin-iso8859-3)
217 (iso-8859-4 latin-iso8859-4)
218 (iso-8859-5 cyrillic-iso8859-5)
219 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
220 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
221 ;; charset is koi8-r, not iso-8859-5.
222 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
223 (iso-8859-6 arabic-iso8859-6)
224 (iso-8859-7 greek-iso8859-7)
225 (iso-8859-8 hebrew-iso8859-8)
226 (iso-8859-9 latin-iso8859-9)
227 (iso-8859-14 latin-iso8859-14)
228 (iso-8859-15 latin-iso8859-15)
229 (viscii vietnamese-viscii-lower)
230 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
231 (euc-kr korean-ksc5601)
232 (gb2312 chinese-gb2312)
233 (big5 chinese-big5-1 chinese-big5-2)
235 (thai-tis620 thai-tis620)
236 (windows-1251 cyrillic-iso8859-5)
237 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
238 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
239 latin-jisx0201 japanese-jisx0208-1978
240 chinese-gb2312 japanese-jisx0208
241 korean-ksc5601 japanese-jisx0212)
242 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
243 latin-jisx0201 japanese-jisx0208-1978
244 chinese-gb2312 japanese-jisx0208
245 korean-ksc5601 japanese-jisx0212
246 chinese-cns11643-1 chinese-cns11643-2)
247 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
248 cyrillic-iso8859-5 greek-iso8859-7
249 latin-jisx0201 japanese-jisx0208-1978
250 chinese-gb2312 japanese-jisx0208
251 korean-ksc5601 japanese-jisx0212
252 chinese-cns11643-1 chinese-cns11643-2
253 chinese-cns11643-3 chinese-cns11643-4
254 chinese-cns11643-5 chinese-cns11643-6
256 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
257 japanese-jisx0213-1 japanese-jisx0213-2)
258 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
259 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
260 (charsetp 'unicode-a)
261 (not (mm-coding-system-p 'mule-utf-8)))
262 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
263 ;; If we have utf-8 we're in Mule 5+.
266 (coding-system-get 'mule-utf-8 'safe-charsets)))))
267 "Alist of MIME-charset/MULE-charsets.")
269 (defun mm-enrich-utf-8-by-mule-ucs ()
270 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
271 This function will run when the `un-define' module is loaded under
272 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
273 with Mule charsets. It is completely useless for Emacs."
274 (unless (cdr (delete '(mm-enrich-utf-8-by-mule-ucs)
275 (assoc "un-define" after-load-alist)))
276 (setq after-load-alist
277 (delete '("un-define") after-load-alist)))
278 (when (boundp 'unicode-basic-translation-charset-order-list)
284 'unicode-basic-translation-charset-order-list))))
285 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
288 (setq mm-mime-mule-charset-alist
289 (nconc mm-mime-mule-charset-alist
290 (list (cons 'utf-8 val))))))
293 ;; Correct by construction, but should be unnecessary for Emacs:
294 (if (featurep 'xemacs)
295 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
296 (when (and (fboundp 'coding-system-list)
297 (fboundp 'sort-coding-systems))
298 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
302 mime (or (coding-system-get cs :mime-charset) ; Emacs 22
303 (coding-system-get cs 'mime-charset)))
305 (not (eq t (setq mule
306 (coding-system-get cs 'safe-charsets))))
307 (not (assq mime alist)))
308 (push (cons mime (delq 'ascii mule)) alist)))
309 (setq mm-mime-mule-charset-alist (nreverse alist)))))
311 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
312 "A list of special charsets.
313 Valid elements include:
314 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
315 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
318 (defvar mm-iso-8859-15-compatible
319 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
320 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
321 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
323 (defvar mm-iso-8859-x-to-15-table
324 (and (fboundp 'coding-system-p)
325 (mm-coding-system-p 'iso-8859-15)
328 (if (mm-coding-system-p (car cs))
329 (let ((c (string-to-char
330 (decode-coding-string "\341" (car cs)))))
331 (cons (char-charset c)
334 (decode-coding-string "\341" 'iso-8859-15)) c)
335 (string-to-list (decode-coding-string (car (cdr cs))
338 mm-iso-8859-15-compatible))
339 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
341 (defcustom mm-coding-system-priorities
342 (if (boundp 'current-language-environment)
343 (let ((lang (symbol-value 'current-language-environment)))
344 (cond ((string= lang "Japanese")
345 ;; Japanese users prefer iso-2022-jp to euc-japan or
346 ;; shift_jis, however iso-8859-1 should be used when
347 ;; there are only ASCII text and Latin-1 characters.
348 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
349 "Preferred coding systems for encoding outgoing messages.
351 More than one suitable coding system may be found for some text.
352 By default, the coding system with the highest priority is used
353 to encode outgoing messages (see `sort-coding-systems'). If this
354 variable is set, it overrides the default priority."
356 :type '(repeat (symbol :tag "Coding system"))
360 (defvar mm-use-find-coding-systems-region
361 (fboundp 'find-coding-systems-region)
362 "Use `find-coding-systems-region' to find proper coding systems.
364 Setting it to nil is useful on Emacsen supporting Unicode if sending
365 mail with multiple parts is preferred to sending a Unicode one.")
367 ;;; Internal variables:
371 (defun mm-mule-charset-to-mime-charset (charset)
372 "Return the MIME charset corresponding to the given Mule CHARSET."
373 (if (and (fboundp 'find-coding-systems-for-charsets)
374 (fboundp 'sort-coding-systems))
375 (let ((css (sort (sort-coding-systems
376 (find-coding-systems-for-charsets (list charset)))
377 'mm-sort-coding-systems-predicate))
379 (while (and (not mime)
381 (when (setq cs (pop css))
382 (setq mime (or (coding-system-get cs :mime-charset)
383 (coding-system-get cs 'mime-charset)))))
385 (let ((alist (mapcar (lambda (cs)
386 (assq cs mm-mime-mule-charset-alist))
387 (sort (mapcar 'car mm-mime-mule-charset-alist)
388 'mm-sort-coding-systems-predicate)))
391 (when (memq charset (cdar alist))
392 (setq out (caar alist)
397 (defun mm-charset-to-coding-system (charset &optional lbt)
398 "Return coding-system corresponding to CHARSET.
399 CHARSET is a symbol naming a MIME charset.
400 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
401 used as the line break code type of the coding system."
402 (when (stringp charset)
403 (setq charset (intern (downcase charset))))
405 (setq charset (intern (format "%s-%s" charset lbt))))
409 ;; Running in a non-MULE environment.
410 ((or (null (mm-get-coding-system-list))
411 (not (fboundp 'coding-system-get)))
414 ((eq charset 'us-ascii)
416 ;; Check to see whether we can handle this charset. (This depends
417 ;; on there being some coding system matching each `mime-charset'
418 ;; property defined, as there should be.)
419 ((and (mm-coding-system-p charset)
420 ;;; Doing this would potentially weed out incorrect charsets.
422 ;;; (eq charset (coding-system-get charset 'mime-charset))
425 ;; Translate invalid charsets.
426 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
427 (and cs (mm-coding-system-p cs) cs)))
428 ;; Last resort: search the coding system list for entries which
429 ;; have the right mime-charset in case the canonical name isn't
430 ;; defined (though it should be).
432 ;; mm-get-coding-system-list returns a list of cs without lbt.
434 (dolist (c (mm-get-coding-system-list))
436 (eq charset (or (coding-system-get c :mime-charset)
437 (coding-system-get c 'mime-charset))))
442 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
443 (boundp 'default-enable-multibyte-characters)
444 default-enable-multibyte-characters
445 (fboundp 'set-buffer-multibyte))
446 "True in Emacs with Mule.")
449 (defun mm-enable-multibyte ()
450 "Set the multibyte flag of the current buffer.
451 Only do this if the default value of `enable-multibyte-characters' is
452 non-nil. This is a no-op in XEmacs."
453 (set-buffer-multibyte 'to))
454 (defalias 'mm-enable-multibyte 'ignore))
457 (defun mm-disable-multibyte ()
458 "Unset the multibyte flag of in the current buffer.
459 This is a no-op in XEmacs."
460 (set-buffer-multibyte nil))
461 (defalias 'mm-disable-multibyte 'ignore)))
463 (defun mm-preferred-coding-system (charset)
464 ;; A typo in some Emacs versions.
465 (or (get-charset-property charset 'preferred-coding-system)
466 (get-charset-property charset 'prefered-coding-system)))
468 ;; Mule charsets shouldn't be used.
469 (defsubst mm-guess-charset ()
470 "Guess Mule charset from the language environment."
472 mail-parse-mule-charset ;; cached mule-charset
474 (setq mail-parse-mule-charset
475 (and (boundp 'current-language-environment)
478 (assoc current-language-environment
479 language-info-alist))))))
480 (if (or (not mail-parse-mule-charset)
481 (eq mail-parse-mule-charset 'ascii))
482 (setq mail-parse-mule-charset
483 (or (car (last (assq mail-parse-charset
484 mm-mime-mule-charset-alist)))
487 mail-parse-mule-charset)))
489 (defun mm-charset-after (&optional pos)
490 "Return charset of a character in current buffer at position POS.
491 If POS is nil, it defauls to the current point.
492 If POS is out of range, the value is nil.
493 If the charset is `composition', return the actual one."
494 (let ((char (char-after pos)) charset)
495 (if (< (mm-char-int char) 128)
496 (setq charset 'ascii)
497 ;; charset-after is fake in some Emacsen.
498 (setq charset (and (fboundp 'char-charset) (char-charset char)))
499 (if (eq charset 'composition) ; Mule 4
500 (let ((p (or pos (point))))
501 (cadr (find-charset-region p (1+ p))))
502 (if (and charset (not (memq charset '(ascii eight-bit-control
503 eight-bit-graphic))))
505 (mm-guess-charset))))))
507 (defun mm-mime-charset (charset)
508 "Return the MIME charset corresponding to the given Mule CHARSET."
509 (if (eq charset 'unknown)
510 (error "The message contains non-printable characters, please use attachment"))
511 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
512 ;; This exists in Emacs 20.
514 (and (mm-preferred-coding-system charset)
515 (or (coding-system-get
516 (mm-preferred-coding-system charset) :mime-charset)
518 (mm-preferred-coding-system charset) 'mime-charset)))
519 (and (eq charset 'ascii)
521 (mm-preferred-coding-system charset)
522 (mm-mule-charset-to-mime-charset charset))
523 ;; This is for XEmacs.
524 (mm-mule-charset-to-mime-charset charset)))
526 (defun mm-delete-duplicates (list)
527 "Simple substitute for CL `delete-duplicates', testing with `equal'."
530 (setq head (car list))
531 (setq list (delete head list))
532 (setq result (cons head result)))
535 ;; Fixme: This is used in places when it should be testing the
536 ;; default multibyteness. See mm-default-multibyte-p.
538 (if (and (not (featurep 'xemacs))
539 (boundp 'enable-multibyte-characters))
540 (defun mm-multibyte-p ()
541 "Non-nil if multibyte is enabled in the current buffer."
542 enable-multibyte-characters)
543 (defun mm-multibyte-p () (featurep 'mule))))
545 (defun mm-default-multibyte-p ()
546 "Return non-nil if the session is multibyte.
547 This affects whether coding conversion should be attempted generally."
549 (if (boundp 'default-enable-multibyte-characters)
550 default-enable-multibyte-characters
553 (defun mm-iso-8859-x-to-15-region (&optional b e)
554 (if (fboundp 'char-charset)
555 (let (charset item c inconvertible)
557 (if e (narrow-to-region b e))
558 (goto-char (point-min))
559 (skip-chars-forward "\0-\177")
562 ((not (setq item (assq (char-charset (setq c (char-after)))
563 mm-iso-8859-x-to-15-table)))
565 ((memq c (cdr (cdr item)))
566 (setq inconvertible t)
569 (insert-before-markers (prog1 (+ c (car (cdr item)))
571 (skip-chars-forward "\0-\177")))
572 (not inconvertible))))
574 (defun mm-sort-coding-systems-predicate (a b)
577 ;; Note: invalid entries are dropped silently
578 (and (setq cs (mm-coding-system-p cs))
579 (coding-system-base cs)))
580 mm-coding-system-priorities)))
581 (and (setq a (mm-coding-system-p a))
582 (if (setq b (mm-coding-system-p b))
583 (> (length (memq (coding-system-base a) priorities))
584 (length (memq (coding-system-base b) priorities)))
588 (autoload 'latin-unity-massage-name "latin-unity")
589 (autoload 'latin-unity-maybe-remap "latin-unity")
590 (autoload 'latin-unity-representations-feasible-region "latin-unity")
591 (autoload 'latin-unity-representations-present-region "latin-unity")
592 (defvar latin-unity-coding-systems)
593 (defvar latin-unity-ucs-list))
595 (defun mm-xemacs-find-mime-charset-1 (begin end)
596 "Determine which MIME charset to use to send region as message.
597 This uses the XEmacs-specific latin-unity package to better handle the
598 case where identical characters from diverse ISO-8859-? character sets
599 can be encoded using a single one of the corresponding coding systems.
601 It treats `mm-coding-system-priorities' as the list of preferred
602 coding systems; a useful example setting for this list in Western
603 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
604 to the very standard Latin 1 coding system, and only move to coding
605 systems that are less supported as is necessary to encode the
606 characters that exist in the buffer.
608 Latin Unity doesn't know about those non-ASCII Roman characters that
609 are available in various East Asian character sets. As such, its
610 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
611 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
612 But this is very much a corner case, so don't worry about it."
613 (let ((systems mm-coding-system-priorities) csets psets curset)
615 ;; Load the Latin Unity library, if available.
616 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
617 (require 'latin-unity))
619 ;; Now, can we use it?
620 (if (featurep 'latin-unity)
622 (setq csets (latin-unity-representations-feasible-region begin end)
623 psets (latin-unity-representations-present-region begin end))
627 ;; Pass back the first coding system in the preferred list
628 ;; that can encode the whole region.
629 (dolist (curset systems)
630 (setq curset (latin-unity-massage-name 'buffer-default curset))
632 ;; If the coding system is a universal coding system, then
633 ;; it can certainly encode all the characters in the region.
634 (if (memq curset latin-unity-ucs-list)
635 (throw 'done (list curset)))
637 ;; If a coding system isn't universal, and isn't in
638 ;; the list that latin unity knows about, we can't
639 ;; decide whether to use it here. Leave that until later
640 ;; in `mm-find-mime-charset-region' function, whence we
642 (unless (memq curset latin-unity-coding-systems)
645 ;; Right, we know about this coding system, and it may
646 ;; conceivably be able to encode all the characters in
648 (if (latin-unity-maybe-remap begin end curset csets psets t)
649 (throw 'done (list curset))))
651 ;; Can't encode using anything from the
652 ;; `mm-coding-system-priorities' list.
653 ;; Leave `mm-find-mime-charset' to do most of the work.
656 ;; Right, latin unity isn't available; let `mm-find-charset-region'
657 ;; take its default action, which equally applies to GNU Emacs.
660 (defmacro mm-xemacs-find-mime-charset (begin end)
661 (when (featurep 'xemacs)
662 `(mm-xemacs-find-mime-charset-1 ,begin ,end)))
664 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
665 "Return the MIME charsets needed to encode the region between B and E.
666 nil means ASCII, a single-element list represents an appropriate MIME
667 charset, and a longer list means no appropriate charset."
669 ;; The return possibilities of this function are a mess...
670 (or (and (mm-multibyte-p)
671 mm-use-find-coding-systems-region
672 ;; Find the mime-charset of the most preferred coding
673 ;; system that has one.
674 (let ((systems (find-coding-systems-region b e)))
675 (when mm-coding-system-priorities
677 (sort systems 'mm-sort-coding-systems-predicate)))
678 (setq systems (delq 'compound-text systems))
679 (unless (equal systems '(undecided))
681 (let* ((head (pop systems))
682 (cs (or (coding-system-get head :mime-charset)
683 (coding-system-get head 'mime-charset))))
684 ;; The mime-charset (`x-ctext') of
685 ;; `compound-text' is not in the IANA list. We
686 ;; shouldn't normally use anything here with a
687 ;; mime-charset having an `x-' prefix.
688 ;; Fixme: Allow this to be overridden, since
689 ;; there is existing use of x-ctext.
690 ;; Also people apparently need the coding system
691 ;; `iso-2022-jp-3' (which Mule-UCS defines with
692 ;; mime-charset, though it's not valid).
694 (not (string-match "^[Xx]-" (symbol-name cs)))
695 ;; UTF-16 of any variety is invalid for
696 ;; text parts and, unfortunately, has
697 ;; mime-charset defined both in Mule-UCS
698 ;; and versions of Emacs. (The name
699 ;; might be `mule-utf-16...' or
701 (not (string-match "utf-16" (symbol-name cs))))
703 charsets (list cs))))))
705 ;; If we're XEmacs, and some coding system is appropriate,
706 ;; mm-xemacs-find-mime-charset will return an appropriate list.
707 ;; Otherwise, we'll get nil, and the next setq will get invoked.
708 (setq charsets (mm-xemacs-find-mime-charset b e))
710 ;; We're not multibyte, or a single coding system won't cover it.
712 (mm-delete-duplicates
713 (mapcar 'mm-mime-charset
715 (mm-find-charset-region b e))))))
716 (if (and (> (length charsets) 1)
717 (memq 'iso-8859-15 charsets)
718 (memq 'iso-8859-15 hack-charsets)
719 (save-excursion (mm-iso-8859-x-to-15-region b e)))
720 (mapcar (lambda (x) (setq charsets (delq (car x) charsets)))
721 mm-iso-8859-15-compatible))
722 (if (and (memq 'iso-2022-jp-2 charsets)
723 (memq 'iso-2022-jp-2 hack-charsets))
724 (setq charsets (delq 'iso-2022-jp charsets)))
727 (defmacro mm-with-unibyte-buffer (&rest forms)
728 "Create a temporary buffer, and evaluate FORMS there like `progn'.
729 Use unibyte mode for this."
730 `(let (default-enable-multibyte-characters)
731 (with-temp-buffer ,@forms)))
732 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
733 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
735 (defmacro mm-with-multibyte-buffer (&rest forms)
736 "Create a temporary buffer, and evaluate FORMS there like `progn'.
737 Use multibyte mode for this."
738 `(let ((default-enable-multibyte-characters t))
739 (with-temp-buffer ,@forms)))
740 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
741 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
743 (defmacro mm-with-unibyte-current-buffer (&rest forms)
744 "Evaluate FORMS with current buffer temporarily made unibyte.
745 Also bind `default-enable-multibyte-characters' to nil.
746 Equivalent to `progn' in XEmacs"
747 (let ((multibyte (make-symbol "multibyte"))
748 (buffer (make-symbol "buffer")))
750 (let ((,multibyte enable-multibyte-characters)
751 (,buffer (current-buffer)))
753 (let (default-enable-multibyte-characters)
754 (set-buffer-multibyte nil)
757 (set-buffer-multibyte ,multibyte)))
758 (let (default-enable-multibyte-characters)
760 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
761 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
763 (defmacro mm-with-unibyte (&rest forms)
764 "Eval the FORMS with the default value of `enable-multibyte-characters' nil."
765 `(let (default-enable-multibyte-characters)
767 (put 'mm-with-unibyte 'lisp-indent-function 0)
768 (put 'mm-with-unibyte 'edebug-form-spec '(body))
770 (defmacro mm-with-multibyte (&rest forms)
771 "Eval the FORMS with the default value of `enable-multibyte-characters' t."
772 `(let ((default-enable-multibyte-characters t))
774 (put 'mm-with-multibyte 'lisp-indent-function 0)
775 (put 'mm-with-multibyte 'edebug-form-spec '(body))
777 (defun mm-find-charset-region (b e)
778 "Return a list of Emacs charsets in the region B to E."
780 ((and (mm-multibyte-p)
781 (fboundp 'find-charset-region))
782 ;; Remove composition since the base charsets have been included.
783 ;; Remove eight-bit-*, treat them as ascii.
784 (let ((css (find-charset-region b e)))
785 (mapcar (lambda (cs) (setq css (delq cs css)))
786 '(composition eight-bit-control eight-bit-graphic
790 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
793 (narrow-to-region b e)
794 (goto-char (point-min))
795 (skip-chars-forward "\0-\177")
800 (and (boundp 'current-language-environment)
801 (car (last (assq 'charset
802 (assoc current-language-environment
803 language-info-alist))))))
804 (if (eq charset 'ascii) (setq charset nil))
807 (car (last (assq mail-parse-charset
808 mm-mime-mule-charset-alist)))))
809 (list 'ascii (or charset 'latin-iso8859-1)))))))))
811 (defun mm-auto-mode-alist ()
812 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
813 (let ((alist auto-mode-alist)
816 (when (listp (cdar alist))
817 (push (car alist) out))
821 (defvar mm-inhibit-file-name-handlers
822 '(jka-compr-handler image-file-handler)
823 "A list of handlers doing (un)compression (etc) thingies.")
825 (defun mm-insert-file-contents (filename &optional visit beg end replace
827 "Like `insert-file-contents', but only reads in the file.
828 A buffer may be modified in several ways after reading into the buffer due
829 to advanced Emacs features, such as file-name-handlers, format decoding,
830 `find-file-hooks', etc.
831 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
832 This function ensures that none of these modifications will take place."
833 (let ((format-alist nil)
834 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
835 (default-major-mode 'fundamental-mode)
836 (enable-local-variables nil)
837 (after-insert-file-functions nil)
838 (enable-local-eval nil)
839 (find-file-hooks nil)
840 (inhibit-file-name-operation (if inhibit
841 'insert-file-contents
842 inhibit-file-name-operation))
843 (inhibit-file-name-handlers
845 (append mm-inhibit-file-name-handlers
846 inhibit-file-name-handlers)
847 inhibit-file-name-handlers)))
848 (insert-file-contents filename visit beg end replace)))
850 (defun mm-append-to-file (start end filename &optional codesys inhibit)
851 "Append the contents of the region to the end of file FILENAME.
852 When called from a function, expects three arguments,
853 START, END and FILENAME. START and END are buffer positions
854 saying what text to write.
855 Optional fourth argument specifies the coding system to use when
857 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
858 (let ((coding-system-for-write
859 (or codesys mm-text-coding-system-for-write
860 mm-text-coding-system))
861 (inhibit-file-name-operation (if inhibit
863 inhibit-file-name-operation))
864 (inhibit-file-name-handlers
866 (append mm-inhibit-file-name-handlers
867 inhibit-file-name-handlers)
868 inhibit-file-name-handlers)))
869 (write-region start end filename t 'no-message)
870 (message "Appended to %s" filename)))
872 (defun mm-write-region (start end filename &optional append visit lockname
873 coding-system inhibit)
875 "Like `write-region'.
876 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
877 (let ((coding-system-for-write
878 (or coding-system mm-text-coding-system-for-write
879 mm-text-coding-system))
880 (inhibit-file-name-operation (if inhibit
882 inhibit-file-name-operation))
883 (inhibit-file-name-handlers
885 (append mm-inhibit-file-name-handlers
886 inhibit-file-name-handlers)
887 inhibit-file-name-handlers)))
888 (write-region start end filename append visit lockname)))
890 (defun mm-image-load-path (&optional package)
892 (dolist (path load-path (nreverse result))
895 (setq dir (concat (file-name-directory
896 (directory-file-name path))
897 "etc/" (or package "gnus/")))))
899 (push path result))))
901 ;; Fixme: This doesn't look useful where it's used.
902 (if (fboundp 'detect-coding-region)
903 (defun mm-detect-coding-region (start end)
904 "Like `detect-coding-region' except returning the best one."
905 (let ((coding-systems
906 (detect-coding-region (point) (point-max))))
907 (or (car-safe coding-systems)
909 (defun mm-detect-coding-region (start end)
910 (let ((point (point)))
912 (skip-chars-forward "\0-\177" end)
914 (if (eq (point) end) 'ascii (mm-guess-charset))
915 (goto-char point)))))
917 (if (fboundp 'coding-system-get)
918 (defun mm-detect-mime-charset-region (start end)
919 "Detect MIME charset of the text in the region between START and END."
920 (let ((cs (mm-detect-coding-region start end)))
921 (coding-system-get cs 'mime-charset)))
922 (defun mm-detect-mime-charset-region (start end)
923 "Detect MIME charset of the text in the region between START and END."
924 (let ((cs (mm-detect-coding-region start end)))
930 ;;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
931 ;;; mm-util.el ends here