1 ;;; mm-util.el --- Utility functions for Mule and low level things
3 ;; Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004,
4 ;; 2005, 2006, 2007, 2008 Free Software Foundation, Inc.
6 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
7 ;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
8 ;; This file is part of GNU Emacs.
10 ;; GNU Emacs is free software: you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation, either version 3 of the License, or
13 ;; (at your option) any later version.
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
29 (unless (fboundp 'declare-function) (defmacro declare-function (&rest r))))
31 (eval-when-compile (require 'cl))
35 (if (featurep 'xemacs)
36 (unless (ignore-errors
37 (require 'timer-funcs))
41 (defvar mm-mime-mule-charset-alist )
43 ;; Emulate functions that are not available in every (X)Emacs version.
44 ;; The name of a function is prefixed with mm-, like `mm-char-int' for
45 ;; `char-int' that is a native XEmacs function, not available in Emacs.
46 ;; Gnus programs all should use mm- functions, not the original ones.
50 (let ((nfunc (intern (format "mm-%s" (car elem)))))
51 (if (fboundp (car elem))
52 (defalias nfunc (car elem))
53 (defalias nfunc (cdr elem)))))
54 `(;; `coding-system-list' is not available in XEmacs 21.4 built
55 ;; without the `file-coding' feature.
56 (coding-system-list . ignore)
57 ;; `char-int' is an XEmacs function, not available in Emacs.
59 ;; `coding-system-equal' is an Emacs function, not available in XEmacs.
60 (coding-system-equal . equal)
61 ;; `annotationp' is an XEmacs function, not available in Emacs.
62 (annotationp . ignore)
63 ;; `set-buffer-file-coding-system' is not available in XEmacs 21.4
64 ;; built without the `file-coding' feature.
65 (set-buffer-file-coding-system . ignore)
66 ;; `read-charset' is an Emacs function, not available in XEmacs.
73 (mapcar (lambda (e) (list (symbol-name (car e))))
74 mm-mime-mule-charset-alist)
76 ;; `subst-char-in-string' is not available in XEmacs 21.4.
78 . ,(lambda (from to string &optional inplace)
79 ;; stolen (and renamed) from nnheader.el
80 "Replace characters in STRING from FROM to TO.
81 Unless optional argument INPLACE is non-nil, return a new string."
82 (let ((string (if inplace string (copy-sequence string)))
85 ;; Replace all occurrences of FROM with TO.
87 (when (= (aref string idx) from)
91 ;; `replace-in-string' is an XEmacs function, not available in Emacs.
93 . ,(lambda (string regexp rep &optional literal)
94 "See `replace-regexp-in-string', only the order of args differs."
95 (replace-regexp-in-string regexp rep string nil literal)))
96 ;; `string-as-unibyte' is an Emacs function, not available in XEmacs.
97 (string-as-unibyte . identity)
98 ;; `string-make-unibyte' is an Emacs function, not available in XEmacs.
99 (string-make-unibyte . identity)
100 ;; string-as-multibyte often doesn't really do what you think it does.
102 ;; (aref (string-as-multibyte "\201") 0) -> 129 (aka ?\201)
103 ;; (aref (string-as-multibyte "\300") 0) -> 192 (aka ?\300)
104 ;; (aref (string-as-multibyte "\300\201") 0) -> 192 (aka ?\300)
105 ;; (aref (string-as-multibyte "\300\201") 1) -> 129 (aka ?\201)
107 ;; (aref (string-as-multibyte "\201\300") 0) -> 2240
108 ;; (aref (string-as-multibyte "\201\300") 1) -> <error>
109 ;; Better use string-to-multibyte or encode-coding-string.
110 ;; If you really need string-as-multibyte somewhere it's usually
111 ;; because you're using the internal emacs-mule representation (maybe
112 ;; because you're using string-as-unibyte somewhere), which is
113 ;; generally a problem in itself.
114 ;; Here is an approximate equivalence table to help think about it:
115 ;; (string-as-multibyte s) ~= (decode-coding-string s 'emacs-mule)
116 ;; (string-to-multibyte s) ~= (decode-coding-string s 'binary)
117 ;; (string-make-multibyte s) ~= (decode-coding-string s locale-coding-system)
118 ;; `string-as-multibyte' is an Emacs function, not available in XEmacs.
119 (string-as-multibyte . identity)
120 ;; `multibyte-string-p' is an Emacs function, not available in XEmacs.
121 (multibyte-string-p . ignore)
122 ;; `insert-byte' is available only in Emacs 23.1 or greater.
123 (insert-byte . insert-char)
124 ;; `multibyte-char-to-unibyte' is an Emacs function, not available
126 (multibyte-char-to-unibyte . identity)
127 ;; `set-buffer-multibyte' is an Emacs function, not available in XEmacs.
128 (set-buffer-multibyte . ignore)
129 ;; `special-display-p' is an Emacs function, not available in XEmacs.
131 . ,(lambda (buffer-name)
132 "Returns non-nil if a buffer named BUFFER-NAME gets a special frame."
133 (and special-display-function
134 (or (and (member buffer-name special-display-buffer-names) t)
135 (cdr (assoc buffer-name special-display-buffer-names))
137 (dolist (elem special-display-regexps)
139 (string-match elem buffer-name)
143 (string-match (car elem) buffer-name)
144 (throw 'return (cdr elem)))))))))
145 ;; `substring-no-properties' is available only in Emacs 22.1 or greater.
146 (substring-no-properties
147 . ,(lambda (string &optional from to)
148 "Return a substring of STRING, without text properties.
149 It starts at index FROM and ending before TO.
150 TO may be nil or omitted; then the substring runs to the end of STRING.
151 If FROM is nil or omitted, the substring starts at the beginning of STRING.
152 If FROM or TO is negative, it counts from the end.
154 With one argument, just copy STRING without its properties."
155 (setq string (substring string (or from 0) to))
156 (set-text-properties 0 (length string) nil string)
158 ;; `line-number-at-pos' is available only in Emacs 22.1 or greater
161 . ,(lambda (&optional pos)
162 "Return (narrowed) buffer line number at position POS.
163 If POS is nil, use current buffer location.
164 Counting starts at (point-min), so the value refers
165 to the contents of the accessible portion of the buffer."
166 (let ((opoint (or pos (point))) start)
168 (goto-char (point-min))
172 (1+ (count-lines start (point))))))))))
174 ;; `decode-coding-string', `encode-coding-string', `decode-coding-region'
175 ;; and `encode-coding-region' are available in Emacs and XEmacs built with
176 ;; the `file-coding' feature, but the XEmacs versions treat nil, that is
177 ;; given as the `coding-system' argument, as the `binary' coding system.
179 (if (featurep 'xemacs)
180 (if (featurep 'file-coding)
182 (defun mm-decode-coding-string (str coding-system)
184 (decode-coding-string str coding-system)
186 (defun mm-encode-coding-string (str coding-system)
188 (encode-coding-string str coding-system)
190 (defun mm-decode-coding-region (start end coding-system)
192 (decode-coding-region start end coding-system)))
193 (defun mm-encode-coding-region (start end coding-system)
195 (encode-coding-region start end coding-system))))
196 (defun mm-decode-coding-string (str coding-system) str)
197 (defun mm-encode-coding-string (str coding-system) str)
198 (defalias 'mm-decode-coding-region 'ignore)
199 (defalias 'mm-encode-coding-region 'ignore))
200 (defalias 'mm-decode-coding-string 'decode-coding-string)
201 (defalias 'mm-encode-coding-string 'encode-coding-string)
202 (defalias 'mm-decode-coding-region 'decode-coding-region)
203 (defalias 'mm-encode-coding-region 'encode-coding-region)))
205 ;; `string-to-multibyte' is available only in Emacs 22.1 or greater.
206 (defalias 'mm-string-to-multibyte
210 ((fboundp 'string-to-multibyte)
211 'string-to-multibyte)
214 "Return a multibyte string with the same individual chars as STRING."
216 (lambda (ch) (mm-string-as-multibyte (char-to-string ch)))
219 ;; `char-or-char-int-p' is an XEmacs function, not available in Emacs.
221 (defalias 'mm-char-or-char-int-p
223 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
224 ((fboundp 'char-valid-p) 'char-valid-p)
227 ;; Fixme: This seems always to be used to read a MIME charset, so it
228 ;; should be re-named and fixed (in Emacs) to offer completion only on
229 ;; proper charset names (base coding systems which have a
230 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
232 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
233 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
234 ;; Actually, there should be an `mm-coding-system-mime-charset'.
236 (defalias 'mm-read-coding-system
238 ((fboundp 'read-coding-system)
239 (if (and (featurep 'xemacs)
240 (<= (string-to-number emacs-version) 21.1))
241 (lambda (prompt &optional default-coding-system)
242 (read-coding-system prompt))
243 'read-coding-system))
244 (t (lambda (prompt &optional default-coding-system)
245 "Prompt the user for a coding system."
247 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
248 mm-mime-mule-charset-alist)))))))
250 (defvar mm-coding-system-list nil)
251 (defun mm-get-coding-system-list ()
252 "Get the coding system list."
253 (or mm-coding-system-list
254 (setq mm-coding-system-list (mm-coding-system-list))))
256 (defun mm-coding-system-p (cs)
257 "Return non-nil if CS is a symbol naming a coding system.
258 In XEmacs, also return non-nil if CS is a coding system object.
259 If CS is available, return CS itself in Emacs, and return a coding
260 system object in XEmacs."
261 (if (fboundp 'find-coding-system)
262 (and cs (find-coding-system cs))
263 (if (fboundp 'coding-system-p)
264 (when (coding-system-p cs)
267 (car (memq cs (mm-get-coding-system-list))))))
269 (defun mm-codepage-setup (number &optional alias)
270 "Create a coding system cpNUMBER.
271 The coding system is created using `codepage-setup'. If ALIAS is
272 non-nil, an alias is created and added to
273 `mm-charset-synonym-alist'. If ALIAS is a string, it's used as
274 the alias. Else windows-NUMBER is used."
276 (let ((completion-ignore-case t)
277 (candidates (if (fboundp 'cp-supported-codepages)
278 (cp-supported-codepages)
279 ;; Removed in Emacs 23 (unicode), so signal an error:
280 (error "`codepage-setup' not present in this Emacs version."))))
281 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
282 nil t nil nil "437"))))
284 (setq alias (if (stringp alias)
286 (intern (format "windows-%s" number)))))
287 (let* ((cp (intern (format "cp%s" number))))
288 (unless (mm-coding-system-p cp)
289 (if (fboundp 'codepage-setup) ; silence compiler
290 (codepage-setup number)
291 (error "`codepage-setup' not present in this Emacs version.")))
293 ;; Don't add alias if setup of cp failed.
294 (mm-coding-system-p cp))
295 (add-to-list 'mm-charset-synonym-alist (cons alias cp)))))
297 (defvar mm-charset-synonym-alist
299 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
300 ,@(unless (mm-coding-system-p 'x-ctext)
301 '((x-ctext . ctext)))
302 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_ in 8
304 ,@(unless (mm-coding-system-p 'iso-8859-15)
305 '((iso-8859-15 . iso-8859-1)))
306 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
307 ,@(unless (mm-coding-system-p 'big5-hkscs)
308 '((big5-hkscs . big5)))
309 ;; A Microsoft misunderstanding.
310 ,@(when (and (not (mm-coding-system-p 'unicode))
311 (mm-coding-system-p 'utf-16-le))
312 '((unicode . utf-16-le)))
313 ;; A Microsoft misunderstanding.
314 ,@(unless (mm-coding-system-p 'ks_c_5601-1987)
315 (if (mm-coding-system-p 'cp949)
316 '((ks_c_5601-1987 . cp949))
317 '((ks_c_5601-1987 . euc-kr))))
318 ;; Windows-31J is Windows Codepage 932.
319 ,@(when (and (not (mm-coding-system-p 'windows-31j))
320 (mm-coding-system-p 'cp932))
321 '((windows-31j . cp932)))
322 ;; Charset name: GBK, Charset aliases: CP936, MS936, windows-936
323 ;; http://www.iana.org/assignments/charset-reg/GBK
324 ;; Emacs 22.1 has cp936, but not gbk, so we alias it:
325 ,@(when (and (not (mm-coding-system-p 'gbk))
326 (mm-coding-system-p 'cp936))
328 ;; UTF8 is a bogus name for UTF-8
329 ,@(when (and (not (mm-coding-system-p 'utf8))
330 (mm-coding-system-p 'utf-8))
332 ;; ISO8859-1 is a bogus name for ISO-8859-1
333 ,@(when (and (not (mm-coding-system-p 'iso8859-1))
334 (mm-coding-system-p 'iso-8859-1))
335 '((iso8859-1 . iso-8859-1)))
336 ;; ISO_8859-1 is a bogus name for ISO-8859-1
337 ,@(when (and (not (mm-coding-system-p 'iso_8859-1))
338 (mm-coding-system-p 'iso-8859-1))
339 '((iso_8859-1 . iso-8859-1)))
341 "A mapping from unknown or invalid charset names to the real charset names.
343 See `mm-codepage-iso-8859-list' and `mm-codepage-ibm-list'.")
345 (defcustom mm-codepage-iso-8859-list
346 (list 1250 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
347 ;; Outlook users in Czech republic. Use this to allow reading of
348 ;; their e-mails. cp1250 should be defined by M-x codepage-setup
350 '(1252 . 1) ;; Windows-1252 is a superset of iso-8859-1 (West
351 ;; Europe). See also `gnus-article-dumbquotes-map'.
352 '(1254 . 9) ;; Windows-1254 is a superset of iso-8859-9 (Turkish).
353 '(1255 . 8));; Windows-1255 is a superset of iso-8859-8 (Hebrew).
354 "A list of Windows codepage numbers and iso-8859 charset numbers.
356 If an element is a number corresponding to a supported windows
357 codepage, appropriate entries to `mm-charset-synonym-alist' are
358 added by `mm-setup-codepage-iso-8859'. An element may also be a
359 cons cell where the car is a codepage number and the cdr is the
360 corresponding number of an iso-8859 charset."
361 :type '(list (set :inline t
362 (const 1250 :tag "Central and East European")
363 (const (1252 . 1) :tag "West European")
364 (const (1254 . 9) :tag "Turkish")
365 (const (1255 . 8) :tag "Hebrew"))
369 (integer :tag "Windows codepage number")
370 (cons (integer :tag "Windows codepage number")
371 (integer :tag "iso-8859 charset number")))))
372 :version "22.1" ;; Gnus 5.10.9
375 (defcustom mm-codepage-ibm-list
376 (list 437 ;; (US etc.)
380 863 ;; (Canadian French)
385 866 ;; (Cyrillic - Russian)
390 ;; In Emacs 23 (unicode), cp... and ibm... are aliases.
391 ;; Cf. http://thread.gmane.org/v9lkng5nwy.fsf@marauder.physik.uni-ulm.de
392 "List of IBM codepage numbers.
394 The codepage mappings slighly differ between IBM and other vendors.
395 See \"ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/IBM/README.TXT\".
397 If an element is a number corresponding to a supported windows
398 codepage, appropriate entries to `mm-charset-synonym-alist' are
399 added by `mm-setup-codepage-ibm'."
400 :type '(list (set :inline t
401 (const 437 :tag "US etc.")
402 (const 860 :tag "Portugal")
403 (const 861 :tag "Iceland")
404 (const 862 :tag "Israel")
405 (const 863 :tag "Canadian French")
406 (const 865 :tag "Nordic")
408 (const 850 :tag "Latin 1")
409 (const 855 :tag "Cyrillic")
410 (const 866 :tag "Cyrillic - Russian")
411 (const 857 :tag "Turkish")
412 (const 864 :tag "Arabic")
413 (const 869 :tag "Greek")
414 (const 874 :tag "Thai"))
417 (integer :tag "Codepage number")))
418 :version "22.1" ;; Gnus 5.10.9
421 (defun mm-setup-codepage-iso-8859 (&optional list)
422 "Add appropriate entries to `mm-charset-synonym-alist'.
423 Unless LIST is given, `mm-codepage-iso-8859-list' is used."
425 (setq list mm-codepage-iso-8859-list))
427 (let (cp windows iso)
429 (setq cp (intern (format "cp%d" (car i)))
430 windows (intern (format "windows-%d" (car i)))
431 iso (intern (format "iso-8859-%d" (cdr i))))
432 (setq cp (intern (format "cp%d" i))
433 windows (intern (format "windows-%d" i))))
434 (unless (mm-coding-system-p windows)
435 (if (mm-coding-system-p cp)
436 (add-to-list 'mm-charset-synonym-alist (cons windows cp))
437 (add-to-list 'mm-charset-synonym-alist (cons windows iso)))))))
439 (defun mm-setup-codepage-ibm (&optional list)
440 "Add appropriate entries to `mm-charset-synonym-alist'.
441 Unless LIST is given, `mm-codepage-ibm-list' is used."
443 (setq list mm-codepage-ibm-list))
444 (dolist (number list)
445 (let ((ibm (intern (format "ibm%d" number)))
446 (cp (intern (format "cp%d" number))))
447 (when (and (not (mm-coding-system-p ibm))
448 (mm-coding-system-p cp))
449 (add-to-list 'mm-charset-synonym-alist (cons ibm cp))))))
452 (mm-setup-codepage-iso-8859)
453 (mm-setup-codepage-ibm)
455 ;; Note: this has to be defined before `mm-charset-to-coding-system'.
456 (defcustom mm-charset-eval-alist
457 (if (featurep 'xemacs)
458 nil ;; I don't know what would be useful for XEmacs.
459 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
460 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
461 (windows-1250 . (mm-codepage-setup 1250 t))
462 (windows-1251 . (mm-codepage-setup 1251 t))
463 (windows-1253 . (mm-codepage-setup 1253 t))
464 (windows-1257 . (mm-codepage-setup 1257 t))))
465 "An alist of (CHARSET . FORM) pairs.
466 If an article is encoded in an unknown CHARSET, FORM is
467 evaluated. This allows to load additional libraries providing
468 charsets on demand. If supported by your Emacs version, you
469 could use `autoload-coding-system' here."
470 :version "22.1" ;; Gnus 5.10.9
471 :type '(list (set :inline t
472 (const (windows-1250 . (mm-codepage-setup 1250 t)))
473 (const (windows-1251 . (mm-codepage-setup 1251 t)))
474 (const (windows-1253 . (mm-codepage-setup 1253 t)))
475 (const (windows-1257 . (mm-codepage-setup 1257 t)))
476 (const (cp850 . (mm-codepage-setup 850 nil))))
479 (cons (symbol :tag "charset")
480 (symbol :tag "form"))))
482 (put 'mm-charset-eval-alist 'risky-local-variable t)
484 (defvar mm-charset-override-alist)
486 ;; Note: this function has to be defined before `mm-charset-override-alist'
487 ;; since it will use this function in order to determine its default value
488 ;; when loading mm-util.elc.
489 (defun mm-charset-to-coding-system (charset &optional lbt
490 allow-override silent)
491 "Return coding-system corresponding to CHARSET.
492 CHARSET is a symbol naming a MIME charset.
493 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
494 used as the line break code type of the coding system.
496 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
497 map undesired charset names to their replacement. This should
498 only be used for decoding, not for encoding.
500 A non-nil value of SILENT means don't issue a warning even if CHARSET
502 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
503 (when (stringp charset)
504 (setq charset (intern (downcase charset))))
506 (setq charset (intern (format "%s-%s" charset lbt))))
510 ;; Running in a non-MULE environment.
511 ((or (null (mm-get-coding-system-list))
512 (not (fboundp 'coding-system-get)))
514 ;; Check override list quite early. Should only used for decoding, not for
517 (let ((cs (cdr (assq charset mm-charset-override-alist))))
518 (and cs (mm-coding-system-p cs) cs))))
520 ((eq charset 'us-ascii)
522 ;; Check to see whether we can handle this charset. (This depends
523 ;; on there being some coding system matching each `mime-charset'
524 ;; property defined, as there should be.)
525 ((and (mm-coding-system-p charset)
526 ;;; Doing this would potentially weed out incorrect charsets.
528 ;;; (eq charset (coding-system-get charset 'mime-charset))
531 ;; Eval expressions from `mm-charset-eval-alist'
532 ((let* ((el (assq charset mm-charset-eval-alist))
539 (condition-case nil (eval form) (error nil))
540 ;; (message "Failed to eval `%s'" form))
541 (mm-coding-system-p cs)
542 (message "Added charset `%s' via `mm-charset-eval-alist'" cs))
544 ;; Translate invalid charsets.
545 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
547 (mm-coding-system-p cs)
549 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
552 ;; Last resort: search the coding system list for entries which
553 ;; have the right mime-charset in case the canonical name isn't
554 ;; defined (though it should be).
556 ;; mm-get-coding-system-list returns a list of cs without lbt.
558 (dolist (c (mm-get-coding-system-list))
560 (eq charset (or (coding-system-get c :mime-charset)
561 (coding-system-get c 'mime-charset))))
563 (unless (or silent cs)
564 ;; Warn the user about unknown charset:
565 (if (fboundp 'gnus-message)
566 (gnus-message 7 "Unknown charset: %s" charset)
567 (message "Unknown charset: %s" charset)))
570 ;; Note: `mm-charset-to-coding-system' has to be defined before this.
571 (defcustom mm-charset-override-alist
572 ;; Note: pairs that cannot be used in the Emacs version currently running
575 (iso-8859-1 . windows-1252)
576 (iso-8859-8 . windows-1255)
577 (iso-8859-9 . windows-1254))
578 "A mapping from undesired charset names to their replacement.
580 You may add pairs like (iso-8859-1 . windows-1252) here,
581 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
582 superset of iso-8859-1."
589 (mapcar (lambda (pair)
590 (if (mm-charset-to-coding-system (cdr pair)
594 (iso-8859-1 . windows-1252)
595 (iso-8859-8 . windows-1255)
596 (iso-8859-9 . windows-1254)
597 (undecided . windows-1252)))))
598 (val (copy-sequence (default-value 'mm-charset-override-alist)))
601 (push (if (and (prog1
602 (setq pair (assq (caar val) defaults))
603 (setq defaults (delq pair defaults)))
604 (equal (car val) pair))
607 (const :format "(%v" ,(caar val))
608 (symbol :size 3 :format " . %v)\n" ,(cdar val))))
610 (setq val (cdr val)))
612 (push `(const ,(pop defaults)) rest))
615 `(set :inline t :format "%v" ,@(nreverse rest))
616 `(repeat :inline t :tag "Other options"
618 (symbol :size 3 :format "(%v")
619 (symbol :size 3 :format " . %v)\n")))))))
620 ;; Remove pairs that cannot be used in the Emacs version currently
621 ;; running. Note that this section will be evaluated when loading
623 :set (lambda (symbol value)
626 (mapcar (lambda (pair)
627 (if (mm-charset-to-coding-system (cdr pair)
631 :version "22.1" ;; Gnus 5.10.9
634 (defvar mm-binary-coding-system
636 ((mm-coding-system-p 'binary) 'binary)
637 ((mm-coding-system-p 'no-conversion) 'no-conversion)
639 "100% binary coding system.")
641 (defvar mm-text-coding-system
642 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
643 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
644 (and (mm-coding-system-p 'raw-text) 'raw-text))
645 mm-binary-coding-system)
646 "Text-safe coding system (For removing ^M).")
648 (defvar mm-text-coding-system-for-write nil
649 "Text coding system for write.")
651 (defvar mm-auto-save-coding-system
653 ((mm-coding-system-p 'utf-8-emacs) ; Mule 7
654 (if (memq system-type '(windows-nt ms-dos ms-windows))
655 (if (mm-coding-system-p 'utf-8-emacs-dos)
656 'utf-8-emacs-dos mm-binary-coding-system)
658 ((mm-coding-system-p 'emacs-mule)
659 (if (memq system-type '(windows-nt ms-dos ms-windows))
660 (if (mm-coding-system-p 'emacs-mule-dos)
661 'emacs-mule-dos mm-binary-coding-system)
663 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
664 (t mm-binary-coding-system))
665 "Coding system of auto save file.")
667 (defvar mm-universal-coding-system mm-auto-save-coding-system
668 "The universal coding system.")
670 ;; Fixme: some of the cars here aren't valid MIME charsets. That
671 ;; should only matter with XEmacs, though.
672 (defvar mm-mime-mule-charset-alist
674 (iso-8859-1 latin-iso8859-1)
675 (iso-8859-2 latin-iso8859-2)
676 (iso-8859-3 latin-iso8859-3)
677 (iso-8859-4 latin-iso8859-4)
678 (iso-8859-5 cyrillic-iso8859-5)
679 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
680 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
681 ;; charset is koi8-r, not iso-8859-5.
682 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
683 (iso-8859-6 arabic-iso8859-6)
684 (iso-8859-7 greek-iso8859-7)
685 (iso-8859-8 hebrew-iso8859-8)
686 (iso-8859-9 latin-iso8859-9)
687 (iso-8859-14 latin-iso8859-14)
688 (iso-8859-15 latin-iso8859-15)
689 (viscii vietnamese-viscii-lower)
690 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
691 (euc-kr korean-ksc5601)
692 (gb2312 chinese-gb2312)
694 (gb18030 gb18030-2-byte
695 gb18030-4-byte-bmp gb18030-4-byte-smp
696 gb18030-4-byte-ext-1 gb18030-4-byte-ext-2)
697 (big5 chinese-big5-1 chinese-big5-2)
699 (thai-tis620 thai-tis620)
700 (windows-1251 cyrillic-iso8859-5)
701 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
702 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
703 latin-jisx0201 japanese-jisx0208-1978
704 chinese-gb2312 japanese-jisx0208
705 korean-ksc5601 japanese-jisx0212)
706 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
707 latin-jisx0201 japanese-jisx0208-1978
708 chinese-gb2312 japanese-jisx0208
709 korean-ksc5601 japanese-jisx0212
710 chinese-cns11643-1 chinese-cns11643-2)
711 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
712 cyrillic-iso8859-5 greek-iso8859-7
713 latin-jisx0201 japanese-jisx0208-1978
714 chinese-gb2312 japanese-jisx0208
715 korean-ksc5601 japanese-jisx0212
716 chinese-cns11643-1 chinese-cns11643-2
717 chinese-cns11643-3 chinese-cns11643-4
718 chinese-cns11643-5 chinese-cns11643-6
720 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
721 japanese-jisx0213-1 japanese-jisx0213-2)
722 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208)
723 ,(cond ((fboundp 'unicode-precedence-list)
724 (cons 'utf-8 (delq 'ascii (mapcar 'charset-name
725 (unicode-precedence-list)))))
726 ((or (not (fboundp 'charsetp)) ;; non-Mule case
727 (charsetp 'unicode-a)
728 (not (mm-coding-system-p 'mule-utf-8)))
729 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
730 (t ;; If we have utf-8 we're in Mule 5+.
733 (coding-system-get 'mule-utf-8 'safe-charsets))))))
734 "Alist of MIME-charset/MULE-charsets.")
736 (defun mm-enrich-utf-8-by-mule-ucs ()
737 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
738 This function will run when the `un-define' module is loaded under
739 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
740 with Mule charsets. It is completely useless for Emacs."
741 (when (boundp 'unicode-basic-translation-charset-order-list)
747 'unicode-basic-translation-charset-order-list))))
748 (elem (assq 'utf-8 mm-mime-mule-charset-alist)))
751 (setq mm-mime-mule-charset-alist
752 (nconc mm-mime-mule-charset-alist
753 (list (cons 'utf-8 val))))))
756 ;; Correct by construction, but should be unnecessary for Emacs:
757 (if (featurep 'xemacs)
758 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
759 (when (and (fboundp 'coding-system-list)
760 (fboundp 'sort-coding-systems))
761 (let ((css (sort-coding-systems (coding-system-list 'base-only)))
765 mime (or (coding-system-get cs :mime-charset); Emacs 23 (unicode)
766 (coding-system-get cs 'mime-charset)))
768 (not (eq t (setq mule
769 (coding-system-get cs 'safe-charsets))))
770 (not (assq mime alist)))
771 (push (cons mime (delq 'ascii mule)) alist)))
772 (setq mm-mime-mule-charset-alist (nreverse alist)))))
774 (defvar mm-hack-charsets '(iso-8859-15 iso-2022-jp-2)
775 "A list of special charsets.
776 Valid elements include:
777 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
778 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
781 (defvar mm-iso-8859-15-compatible
782 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
783 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
784 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
786 (defvar mm-iso-8859-x-to-15-table
787 (and (fboundp 'coding-system-p)
788 (mm-coding-system-p 'iso-8859-15)
791 (if (mm-coding-system-p (car cs))
792 (let ((c (string-to-char
793 (decode-coding-string "\341" (car cs)))))
794 (cons (char-charset c)
797 (decode-coding-string "\341" 'iso-8859-15)) c)
798 (string-to-list (decode-coding-string (car (cdr cs))
801 mm-iso-8859-15-compatible))
802 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
804 (defcustom mm-coding-system-priorities
805 (if (boundp 'current-language-environment)
806 (let ((lang (symbol-value 'current-language-environment)))
807 (cond ((string= lang "Japanese")
808 ;; Japanese users prefer iso-2022-jp to euc-japan or
809 ;; shift_jis, however iso-8859-1 should be used when
810 ;; there are only ASCII text and Latin-1 characters.
811 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8)))))
812 "Preferred coding systems for encoding outgoing messages.
814 More than one suitable coding system may be found for some text.
815 By default, the coding system with the highest priority is used
816 to encode outgoing messages (see `sort-coding-systems'). If this
817 variable is set, it overrides the default priority."
819 :type '(repeat (symbol :tag "Coding system"))
823 (defvar mm-use-find-coding-systems-region
824 (fboundp 'find-coding-systems-region)
825 "Use `find-coding-systems-region' to find proper coding systems.
827 Setting it to nil is useful on Emacsen supporting Unicode if sending
828 mail with multiple parts is preferred to sending a Unicode one.")
830 ;;; Internal variables:
834 (defun mm-mule-charset-to-mime-charset (charset)
835 "Return the MIME charset corresponding to the given Mule CHARSET."
836 (if (and (fboundp 'find-coding-systems-for-charsets)
837 (fboundp 'sort-coding-systems))
838 (let ((css (sort (sort-coding-systems
839 (find-coding-systems-for-charsets (list charset)))
840 'mm-sort-coding-systems-predicate))
842 (while (and (not mime)
844 (when (setq cs (pop css))
845 (setq mime (or (coding-system-get cs :mime-charset)
846 (coding-system-get cs 'mime-charset)))))
848 (let ((alist (mapcar (lambda (cs)
849 (assq cs mm-mime-mule-charset-alist))
850 (sort (mapcar 'car mm-mime-mule-charset-alist)
851 'mm-sort-coding-systems-predicate)))
854 (when (memq charset (cdar alist))
855 (setq out (caar alist)
861 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
862 (boundp 'default-enable-multibyte-characters)
863 default-enable-multibyte-characters
864 (fboundp 'set-buffer-multibyte))
865 "True in Emacs with Mule.")
868 (defun mm-enable-multibyte ()
869 "Set the multibyte flag of the current buffer.
870 Only do this if the default value of `enable-multibyte-characters' is
871 non-nil. This is a no-op in XEmacs."
872 (set-buffer-multibyte 'to))
873 (defalias 'mm-enable-multibyte 'ignore))
876 (defun mm-disable-multibyte ()
877 "Unset the multibyte flag of in the current buffer.
878 This is a no-op in XEmacs."
879 (set-buffer-multibyte nil))
880 (defalias 'mm-disable-multibyte 'ignore)))
882 (defun mm-preferred-coding-system (charset)
883 ;; A typo in some Emacs versions.
884 (or (get-charset-property charset 'preferred-coding-system)
885 (get-charset-property charset 'prefered-coding-system)))
887 ;; Mule charsets shouldn't be used.
888 (defsubst mm-guess-charset ()
889 "Guess Mule charset from the language environment."
891 mail-parse-mule-charset ;; cached mule-charset
893 (setq mail-parse-mule-charset
894 (and (boundp 'current-language-environment)
897 (assoc current-language-environment
898 language-info-alist))))))
899 (if (or (not mail-parse-mule-charset)
900 (eq mail-parse-mule-charset 'ascii))
901 (setq mail-parse-mule-charset
902 (or (car (last (assq mail-parse-charset
903 mm-mime-mule-charset-alist)))
906 mail-parse-mule-charset)))
908 (defun mm-charset-after (&optional pos)
909 "Return charset of a character in current buffer at position POS.
910 If POS is nil, it defauls to the current point.
911 If POS is out of range, the value is nil.
912 If the charset is `composition', return the actual one."
913 (let ((char (char-after pos)) charset)
914 (if (< (mm-char-int char) 128)
915 (setq charset 'ascii)
916 ;; charset-after is fake in some Emacsen.
917 (setq charset (and (fboundp 'char-charset) (char-charset char)))
918 (if (eq charset 'composition) ; Mule 4
919 (let ((p (or pos (point))))
920 (cadr (find-charset-region p (1+ p))))
921 (if (and charset (not (memq charset '(ascii eight-bit-control
922 eight-bit-graphic))))
924 (mm-guess-charset))))))
926 (defun mm-mime-charset (charset)
927 "Return the MIME charset corresponding to the given Mule CHARSET."
928 (if (eq charset 'unknown)
929 (error "The message contains non-printable characters, please use attachment"))
930 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
931 ;; This exists in Emacs 20.
933 (and (mm-preferred-coding-system charset)
934 (or (coding-system-get
935 (mm-preferred-coding-system charset) :mime-charset)
937 (mm-preferred-coding-system charset) 'mime-charset)))
938 (and (eq charset 'ascii)
940 (mm-preferred-coding-system charset)
941 (mm-mule-charset-to-mime-charset charset))
942 ;; This is for XEmacs.
943 (mm-mule-charset-to-mime-charset charset)))
945 (if (fboundp 'delete-dups)
946 (defalias 'mm-delete-duplicates 'delete-dups)
947 (defun mm-delete-duplicates (list)
948 "Destructively remove `equal' duplicates from LIST.
949 Store the result in LIST and return it. LIST must be a proper list.
950 Of several `equal' occurrences of an element in LIST, the first
953 This is a compatibility function for Emacsen without `delete-dups'."
954 ;; Code from `subr.el' in Emacs 22:
957 (setcdr tail (delete (car tail) (cdr tail)))
958 (setq tail (cdr tail))))
961 ;; Fixme: This is used in places when it should be testing the
962 ;; default multibyteness. See mm-default-multibyte-p.
964 (if (and (not (featurep 'xemacs))
965 (boundp 'enable-multibyte-characters))
966 (defun mm-multibyte-p ()
967 "Non-nil if multibyte is enabled in the current buffer."
968 enable-multibyte-characters)
969 (defun mm-multibyte-p () (featurep 'mule))))
971 (defun mm-default-multibyte-p ()
972 "Return non-nil if the session is multibyte.
973 This affects whether coding conversion should be attempted generally."
975 (if (boundp 'default-enable-multibyte-characters)
976 default-enable-multibyte-characters
979 (defun mm-iso-8859-x-to-15-region (&optional b e)
980 (if (fboundp 'char-charset)
981 (let (charset item c inconvertible)
983 (if e (narrow-to-region b e))
984 (goto-char (point-min))
985 (skip-chars-forward "\0-\177")
988 ((not (setq item (assq (char-charset (setq c (char-after)))
989 mm-iso-8859-x-to-15-table)))
991 ((memq c (cdr (cdr item)))
992 (setq inconvertible t)
995 (insert-before-markers (prog1 (+ c (car (cdr item)))
997 (skip-chars-forward "\0-\177")))
998 (not inconvertible))))
1000 (defun mm-sort-coding-systems-predicate (a b)
1002 (mapcar (lambda (cs)
1003 ;; Note: invalid entries are dropped silently
1004 (and (setq cs (mm-coding-system-p cs))
1005 (coding-system-base cs)))
1006 mm-coding-system-priorities)))
1007 (and (setq a (mm-coding-system-p a))
1008 (if (setq b (mm-coding-system-p b))
1009 (> (length (memq (coding-system-base a) priorities))
1010 (length (memq (coding-system-base b) priorities)))
1014 (autoload 'latin-unity-massage-name "latin-unity")
1015 (autoload 'latin-unity-maybe-remap "latin-unity")
1016 (autoload 'latin-unity-representations-feasible-region "latin-unity")
1017 (autoload 'latin-unity-representations-present-region "latin-unity"))
1019 (defvar latin-unity-coding-systems)
1020 (defvar latin-unity-ucs-list)
1022 (defun mm-xemacs-find-mime-charset-1 (begin end)
1023 "Determine which MIME charset to use to send region as message.
1024 This uses the XEmacs-specific latin-unity package to better handle the
1025 case where identical characters from diverse ISO-8859-? character sets
1026 can be encoded using a single one of the corresponding coding systems.
1028 It treats `mm-coding-system-priorities' as the list of preferred
1029 coding systems; a useful example setting for this list in Western
1030 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
1031 to the very standard Latin 1 coding system, and only move to coding
1032 systems that are less supported as is necessary to encode the
1033 characters that exist in the buffer.
1035 Latin Unity doesn't know about those non-ASCII Roman characters that
1036 are available in various East Asian character sets. As such, its
1037 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
1038 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
1039 But this is very much a corner case, so don't worry about it."
1040 (let ((systems mm-coding-system-priorities) csets psets curset)
1042 ;; Load the Latin Unity library, if available.
1043 (when (and (not (featurep 'latin-unity)) (locate-library "latin-unity"))
1044 (require 'latin-unity))
1046 ;; Now, can we use it?
1047 (if (featurep 'latin-unity)
1049 (setq csets (latin-unity-representations-feasible-region begin end)
1050 psets (latin-unity-representations-present-region begin end))
1054 ;; Pass back the first coding system in the preferred list
1055 ;; that can encode the whole region.
1056 (dolist (curset systems)
1057 (setq curset (latin-unity-massage-name 'buffer-default curset))
1059 ;; If the coding system is a universal coding system, then
1060 ;; it can certainly encode all the characters in the region.
1061 (if (memq curset latin-unity-ucs-list)
1062 (throw 'done (list curset)))
1064 ;; If a coding system isn't universal, and isn't in
1065 ;; the list that latin unity knows about, we can't
1066 ;; decide whether to use it here. Leave that until later
1067 ;; in `mm-find-mime-charset-region' function, whence we
1068 ;; have been called.
1069 (unless (memq curset latin-unity-coding-systems)
1072 ;; Right, we know about this coding system, and it may
1073 ;; conceivably be able to encode all the characters in
1075 (if (latin-unity-maybe-remap begin end curset csets psets t)
1076 (throw 'done (list curset))))
1078 ;; Can't encode using anything from the
1079 ;; `mm-coding-system-priorities' list.
1080 ;; Leave `mm-find-mime-charset' to do most of the work.
1083 ;; Right, latin unity isn't available; let `mm-find-charset-region'
1084 ;; take its default action, which equally applies to GNU Emacs.
1087 (defmacro mm-xemacs-find-mime-charset (begin end)
1088 (when (featurep 'xemacs)
1089 `(and (featurep 'mule) (mm-xemacs-find-mime-charset-1 ,begin ,end))))
1091 (declare-function mm-delete-duplicates "mm-util" (list))
1093 (defun mm-find-mime-charset-region (b e &optional hack-charsets)
1094 "Return the MIME charsets needed to encode the region between B and E.
1095 nil means ASCII, a single-element list represents an appropriate MIME
1096 charset, and a longer list means no appropriate charset."
1098 ;; The return possibilities of this function are a mess...
1099 (or (and (mm-multibyte-p)
1100 mm-use-find-coding-systems-region
1101 ;; Find the mime-charset of the most preferred coding
1102 ;; system that has one.
1103 (let ((systems (find-coding-systems-region b e)))
1104 (when mm-coding-system-priorities
1106 (sort systems 'mm-sort-coding-systems-predicate)))
1107 (setq systems (delq 'compound-text systems))
1108 (unless (equal systems '(undecided))
1110 (let* ((head (pop systems))
1111 (cs (or (coding-system-get head :mime-charset)
1112 (coding-system-get head 'mime-charset))))
1113 ;; The mime-charset (`x-ctext') of
1114 ;; `compound-text' is not in the IANA list. We
1115 ;; shouldn't normally use anything here with a
1116 ;; mime-charset having an `x-' prefix.
1117 ;; Fixme: Allow this to be overridden, since
1118 ;; there is existing use of x-ctext.
1119 ;; Also people apparently need the coding system
1120 ;; `iso-2022-jp-3' (which Mule-UCS defines with
1121 ;; mime-charset, though it's not valid).
1123 (not (string-match "^[Xx]-" (symbol-name cs)))
1124 ;; UTF-16 of any variety is invalid for
1125 ;; text parts and, unfortunately, has
1126 ;; mime-charset defined both in Mule-UCS
1127 ;; and versions of Emacs. (The name
1128 ;; might be `mule-utf-16...' or
1130 (not (string-match "utf-16" (symbol-name cs))))
1132 charsets (list cs))))))
1134 ;; If we're XEmacs, and some coding system is appropriate,
1135 ;; mm-xemacs-find-mime-charset will return an appropriate list.
1136 ;; Otherwise, we'll get nil, and the next setq will get invoked.
1137 (setq charsets (mm-xemacs-find-mime-charset b e))
1139 ;; Fixme: won't work for unibyte Emacs 23:
1141 ;; We're not multibyte, or a single coding system won't cover it.
1143 (mm-delete-duplicates
1144 (mapcar 'mm-mime-charset
1146 (mm-find-charset-region b e))))))
1147 (if (and (> (length charsets) 1)
1148 (memq 'iso-8859-15 charsets)
1149 (memq 'iso-8859-15 hack-charsets)
1150 (save-excursion (mm-iso-8859-x-to-15-region b e)))
1151 (dolist (x mm-iso-8859-15-compatible)
1152 (setq charsets (delq (car x) charsets))))
1153 (if (and (memq 'iso-2022-jp-2 charsets)
1154 (memq 'iso-2022-jp-2 hack-charsets))
1155 (setq charsets (delq 'iso-2022-jp charsets)))
1156 ;; Attempt to reduce the number of charsets if utf-8 is available.
1157 (if (and (featurep 'xemacs)
1158 (> (length charsets) 1)
1159 (mm-coding-system-p 'utf-8))
1160 (let ((mm-coding-system-priorities
1161 (cons 'utf-8 mm-coding-system-priorities)))
1163 (mm-delete-duplicates
1164 (mapcar 'mm-mime-charset
1166 (mm-find-charset-region b e)))))))
1169 (defmacro mm-with-unibyte-buffer (&rest forms)
1170 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1171 Use unibyte mode for this."
1173 (mm-disable-multibyte)
1175 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
1176 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
1178 (defmacro mm-with-multibyte-buffer (&rest forms)
1179 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1180 Use multibyte mode for this."
1182 (mm-enable-multibyte)
1184 (put 'mm-with-multibyte-buffer 'lisp-indent-function 0)
1185 (put 'mm-with-multibyte-buffer 'edebug-form-spec '(body))
1187 (defmacro mm-with-unibyte-current-buffer (&rest forms)
1188 "Evaluate FORMS with current buffer temporarily made unibyte.
1189 Also bind `default-enable-multibyte-characters' to nil.
1190 Equivalent to `progn' in XEmacs
1192 NOTE: Use this macro with caution in multibyte buffers (it is not
1193 worth using this macro in unibyte buffers of course). Use of
1194 `(set-buffer-multibyte t)', which is run finally, is generally
1195 harmful since it is likely to modify existing data in the buffer.
1196 For instance, it converts \"\\300\\255\" into \"\\255\" in
1197 Emacs 23 (unicode)."
1198 (let ((multibyte (make-symbol "multibyte"))
1199 (buffer (make-symbol "buffer")))
1201 (let ((,multibyte enable-multibyte-characters)
1202 (,buffer (current-buffer)))
1204 (let (default-enable-multibyte-characters)
1205 (set-buffer-multibyte nil)
1207 (set-buffer ,buffer)
1208 (set-buffer-multibyte ,multibyte)))
1209 (let (default-enable-multibyte-characters)
1211 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
1212 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
1214 (defun mm-find-charset-region (b e)
1215 "Return a list of Emacs charsets in the region B to E."
1217 ((and (mm-multibyte-p)
1218 (fboundp 'find-charset-region))
1219 ;; Remove composition since the base charsets have been included.
1220 ;; Remove eight-bit-*, treat them as ascii.
1221 (let ((css (find-charset-region b e)))
1223 '(composition eight-bit-control eight-bit-graphic control-1)
1225 (setq css (delq cs css)))))
1227 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
1230 (narrow-to-region b e)
1231 (goto-char (point-min))
1232 (skip-chars-forward "\0-\177")
1237 (and (boundp 'current-language-environment)
1238 (car (last (assq 'charset
1239 (assoc current-language-environment
1240 language-info-alist))))))
1241 (if (eq charset 'ascii) (setq charset nil))
1244 (car (last (assq mail-parse-charset
1245 mm-mime-mule-charset-alist)))))
1246 (list 'ascii (or charset 'latin-iso8859-1)))))))))
1248 (defun mm-auto-mode-alist ()
1249 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
1250 (let ((alist auto-mode-alist)
1253 (when (listp (cdar alist))
1254 (push (car alist) out))
1258 (defvar mm-inhibit-file-name-handlers
1259 '(jka-compr-handler image-file-handler epa-file-handler)
1260 "A list of handlers doing (un)compression (etc) thingies.")
1262 (defun mm-insert-file-contents (filename &optional visit beg end replace
1264 "Like `insert-file-contents', but only reads in the file.
1265 A buffer may be modified in several ways after reading into the buffer due
1266 to advanced Emacs features, such as file-name-handlers, format decoding,
1267 `find-file-hooks', etc.
1268 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
1269 This function ensures that none of these modifications will take place."
1270 (let* ((format-alist nil)
1271 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
1272 (default-major-mode 'fundamental-mode)
1273 (enable-local-variables nil)
1274 (after-insert-file-functions nil)
1275 (enable-local-eval nil)
1276 (inhibit-file-name-operation (if inhibit
1277 'insert-file-contents
1278 inhibit-file-name-operation))
1279 (inhibit-file-name-handlers
1281 (append mm-inhibit-file-name-handlers
1282 inhibit-file-name-handlers)
1283 inhibit-file-name-handlers))
1284 (ffh (if (boundp 'find-file-hook)
1287 (val (symbol-value ffh)))
1290 (insert-file-contents filename visit beg end replace)
1293 (defun mm-append-to-file (start end filename &optional codesys inhibit)
1294 "Append the contents of the region to the end of file FILENAME.
1295 When called from a function, expects three arguments,
1296 START, END and FILENAME. START and END are buffer positions
1297 saying what text to write.
1298 Optional fourth argument specifies the coding system to use when
1300 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1301 (let ((coding-system-for-write
1302 (or codesys mm-text-coding-system-for-write
1303 mm-text-coding-system))
1304 (inhibit-file-name-operation (if inhibit
1306 inhibit-file-name-operation))
1307 (inhibit-file-name-handlers
1309 (append mm-inhibit-file-name-handlers
1310 inhibit-file-name-handlers)
1311 inhibit-file-name-handlers)))
1312 (write-region start end filename t 'no-message)
1313 (message "Appended to %s" filename)))
1315 (defun mm-write-region (start end filename &optional append visit lockname
1316 coding-system inhibit)
1318 "Like `write-region'.
1319 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1320 (let ((coding-system-for-write
1321 (or coding-system mm-text-coding-system-for-write
1322 mm-text-coding-system))
1323 (inhibit-file-name-operation (if inhibit
1325 inhibit-file-name-operation))
1326 (inhibit-file-name-handlers
1328 (append mm-inhibit-file-name-handlers
1329 inhibit-file-name-handlers)
1330 inhibit-file-name-handlers)))
1331 (write-region start end filename append visit lockname)))
1333 (autoload 'gmm-write-region "gmm-utils")
1335 ;; It is not a MIME function, but some MIME functions use it.
1336 (if (and (fboundp 'make-temp-file)
1338 (let ((def (symbol-function 'make-temp-file)))
1339 (and (byte-code-function-p def)
1340 (setq def (if (fboundp 'compiled-function-arglist)
1342 (eval (list 'compiled-function-arglist def))
1345 (eq (nth 3 def) 'suffix)))))
1346 (defalias 'mm-make-temp-file 'make-temp-file)
1347 ;; Stolen (and modified for XEmacs) from Emacs 22.
1348 (defun mm-make-temp-file (prefix &optional dir-flag suffix)
1349 "Create a temporary file.
1350 The returned file name (created by appending some random characters at the end
1351 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1352 is guaranteed to point to a newly created empty file.
1353 You can then use `write-region' to write new data into the file.
1355 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1357 If SUFFIX is non-nil, add that at the end of the file name."
1358 (let ((umask (default-file-modes))
1362 ;; Create temp files with strict access rights. It's easy to
1363 ;; loosen them later, whereas it's impossible to close the
1364 ;; time-window of loose permissions otherwise.
1365 (set-default-file-modes 448)
1366 (while (condition-case err
1372 (if (fboundp 'temp-directory)
1375 temporary-file-directory))))
1377 (setq file (concat file suffix)))
1379 (make-directory file)
1380 ;; NOTE: This is unsafe if Emacs 20
1381 ;; users and XEmacs users don't use
1382 ;; a secure temp directory.
1383 (gmm-write-region "" nil file nil 'silent
1386 (file-already-exists t)
1387 ;; The XEmacs version of `make-directory' issues
1389 (file-error (or (and (featurep 'xemacs)
1390 (file-exists-p file))
1391 (signal (car err) (cdr err)))))
1392 ;; the file was somehow created by someone else between
1393 ;; `make-temp-name' and `write-region', let's try again.
1397 (set-default-file-modes umask)))))
1399 (defun mm-image-load-path (&optional package)
1401 (dolist (path load-path (nreverse result))
1404 (setq dir (concat (file-name-directory
1405 (directory-file-name path))
1406 "etc/images/" (or package "gnus/")))))
1408 (push path result))))
1410 ;; Fixme: This doesn't look useful where it's used.
1411 (if (fboundp 'detect-coding-region)
1412 (defun mm-detect-coding-region (start end)
1413 "Like `detect-coding-region' except returning the best one."
1414 (let ((coding-systems
1415 (detect-coding-region start end)))
1416 (or (car-safe coding-systems)
1418 (defun mm-detect-coding-region (start end)
1419 (let ((point (point)))
1421 (skip-chars-forward "\0-\177" end)
1423 (if (eq (point) end) 'ascii (mm-guess-charset))
1424 (goto-char point)))))
1426 (declare-function mm-detect-coding-region "mm-util" (start end))
1428 (if (fboundp 'coding-system-get)
1429 (defun mm-detect-mime-charset-region (start end)
1430 "Detect MIME charset of the text in the region between START and END."
1431 (let ((cs (mm-detect-coding-region start end)))
1432 (or (coding-system-get cs :mime-charset)
1433 (coding-system-get cs 'mime-charset))))
1434 (defun mm-detect-mime-charset-region (start end)
1435 "Detect MIME charset of the text in the region between START and END."
1436 (let ((cs (mm-detect-coding-region start end)))
1440 (unless (fboundp 'coding-system-to-mime-charset)
1441 (defalias 'coding-system-to-mime-charset 'ignore)))
1443 (defun mm-coding-system-to-mime-charset (coding-system)
1444 "Return the MIME charset corresponding to CODING-SYSTEM.
1445 To make this function work with XEmacs, the APEL package is required."
1447 (or (and (fboundp 'coding-system-get)
1448 (or (coding-system-get coding-system :mime-charset)
1449 (coding-system-get coding-system 'mime-charset)))
1450 (and (featurep 'xemacs)
1451 (or (and (fboundp 'coding-system-to-mime-charset)
1452 (not (eq (symbol-function 'coding-system-to-mime-charset)
1454 (and (condition-case nil
1457 (fboundp 'coding-system-to-mime-charset)))
1458 (coding-system-to-mime-charset coding-system)))))
1461 (require 'jka-compr))
1463 (defun mm-decompress-buffer (filename &optional inplace force)
1464 "Decompress buffer's contents, depending on jka-compr.
1465 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1466 agrees with `jka-compr-compression-info-list', decompression is done.
1467 Signal an error if FORCE is neither nil nor t and compressed data are
1468 not decompressed because `auto-compression-mode' is disabled.
1469 If INPLACE is nil, return decompressed data or nil without modifying
1470 the buffer. Otherwise, replace the buffer's contents with the
1471 decompressed data. The buffer's multibyteness must be turned off."
1474 (prog1 t (require 'jka-compr))
1475 (and (fboundp 'jka-compr-installed-p)
1476 (jka-compr-installed-p))))
1477 (let ((info (jka-compr-get-compression-info filename)))
1479 (unless (or (memq force (list nil t))
1480 (jka-compr-installed-p))
1482 (let ((prog (jka-compr-info-uncompress-program info))
1483 (args (jka-compr-info-uncompress-args info))
1484 (msg (format "%s %s..."
1485 (jka-compr-info-uncompress-message info)
1487 (err-file (jka-compr-make-temp-name))
1488 (cur (current-buffer))
1489 (coding-system-for-read mm-binary-coding-system)
1490 (coding-system-for-write mm-binary-coding-system)
1493 (mm-with-unibyte-buffer
1494 (insert-buffer-substring cur)
1497 (unless (memq (apply 'call-process-region
1498 (point-min) (point-max)
1499 prog t (list t err-file) nil args)
1500 jka-compr-acceptable-retval-list)
1504 (delete "" (split-string
1506 (insert-file-contents err-file)
1512 (format "Error while executing \"%s %s < %s\""
1513 prog (mapconcat 'identity args " ")
1515 (setq retval (buffer-string)))
1517 (setq err-msg (error-message-string err)))))
1518 (when (file-exists-p err-file)
1519 (ignore-errors (jka-compr-delete-temp-file err-file)))
1522 (delete-region (point-min) (point-max))
1525 (message "%s" (or err-msg (concat msg "done")))
1529 (unless (fboundp 'coding-system-name)
1530 (defalias 'coding-system-name 'ignore))
1531 (unless (fboundp 'find-file-coding-system-for-read-from-filename)
1532 (defalias 'find-file-coding-system-for-read-from-filename 'ignore))
1533 (unless (fboundp 'find-operation-coding-system)
1534 (defalias 'find-operation-coding-system 'ignore)))
1536 (defun mm-find-buffer-file-coding-system (&optional filename)
1537 "Find coding system used to decode the contents of the current buffer.
1538 This function looks for the coding system magic cookie or examines the
1539 coding system specified by `file-coding-system-alist' being associated
1540 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1541 gzip, bzip2, etc. are allowed."
1543 (setq filename buffer-file-name))
1545 (let ((decomp (unless ;; No worth to examine charset of tar files.
1548 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1550 (mm-decompress-buffer filename nil t))))
1552 (set-buffer (let (default-enable-multibyte-characters)
1553 (generate-new-buffer " *temp*")))
1555 (setq filename (file-name-sans-extension filename)))
1556 (goto-char (point-min))
1559 ((boundp 'set-auto-coding-function) ;; Emacs
1561 (or (funcall (symbol-value 'set-auto-coding-function)
1562 filename (- (point-max) (point-min)))
1563 (car (find-operation-coding-system 'insert-file-contents
1565 (let (auto-coding-alist)
1567 (funcall (symbol-value 'set-auto-coding-function)
1568 nil (- (point-max) (point-min)))
1570 ((and (featurep 'xemacs) (featurep 'file-coding)) ;; XEmacs
1571 (let ((case-fold-search t)
1572 (end (point-at-eol))
1575 (and (re-search-forward "-\\*-+[\t ]*" end t)
1577 (setq start (match-end 0))
1578 (re-search-forward "[\t ]*-+\\*-" end t))
1580 (setq end (match-beginning 0))
1582 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1584 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1586 (find-coding-system (setq codesys
1587 (intern (match-string 1))))
1589 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1592 (setq start (match-end 0))
1593 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t))
1595 (setq end (match-beginning 0))
1598 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1600 (find-coding-system (setq codesys
1601 (intern (match-string 1))))
1604 (goto-char (point-min))
1605 (setq case-fold-search nil)
1606 (re-search-forward "^;;;coding system: "
1607 ;;(+ (point-min) 3000) t))
1609 (looking-at "[^\t\n\r ]+")
1611 (setq codesys (intern (match-string 0))))
1615 (find-file-coding-system-for-read-from-filename
1617 (coding-system-name (coding-system-base codesys)))))))
1619 (kill-buffer (current-buffer)))))))
1623 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1624 ;;; mm-util.el ends here