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, 2009 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 ;; `ucs-to-char' is a function that Mule-UCS provides.
228 (if (featurep 'xemacs
)
229 (cond ((and (fboundp 'unicode-to-char
) ;; XEmacs 21.5.
230 (subrp (symbol-function 'unicode-to-char
)))
232 (defalias 'mm-ucs-to-char
'unicode-to-char
)
233 (defun mm-ucs-to-char (codepoint)
234 "Convert Unicode codepoint to character."
235 (or (unicode-to-char codepoint
) ?
#))))
237 (defun mm-ucs-to-char (codepoint)
238 "Convert Unicode codepoint to character."
239 (if (fboundp 'ucs-to-char
) ;; Mule-UCS is loaded.
241 (defalias 'mm-ucs-to-char
243 "Convert Unicode codepoint to character."
245 (or (ucs-to-char codepoint
) ?
#)
247 (mm-ucs-to-char codepoint
))
249 (or (int-to-char codepoint
) ?
#)
252 (defun mm-ucs-to-char (codepoint)
253 "Convert Unicode codepoint to character."
255 (or (int-to-char codepoint
) ?
#)
257 (if (let ((char (make-char 'japanese-jisx0208
36 34)))
258 (eq char
(decode-char 'ucs char
)))
260 (defalias 'mm-ucs-to-char
'identity
)
261 (defun mm-ucs-to-char (codepoint)
262 "Convert Unicode codepoint to character."
263 (or (decode-char 'ucs codepoint
) ?
#))))
265 ;; Fixme: This seems always to be used to read a MIME charset, so it
266 ;; should be re-named and fixed (in Emacs) to offer completion only on
267 ;; proper charset names (base coding systems which have a
268 ;; mime-charset defined). XEmacs doesn't believe in mime-charset;
270 ;; `(or (coding-system-get 'iso-8859-1 'mime-charset)
271 ;; (coding-system-get 'iso-8859-1 :mime-charset))'
272 ;; Actually, there should be an `mm-coding-system-mime-charset'.
274 (defalias 'mm-read-coding-system
276 ((fboundp 'read-coding-system
)
277 (if (and (featurep 'xemacs
)
278 (<= (string-to-number emacs-version
) 21.1))
279 (lambda (prompt &optional default-coding-system
)
280 (read-coding-system prompt
))
281 'read-coding-system
))
282 (t (lambda (prompt &optional default-coding-system
)
283 "Prompt the user for a coding system."
285 prompt
(mapcar (lambda (s) (list (symbol-name (car s
))))
286 mm-mime-mule-charset-alist
)))))))
288 (defvar mm-coding-system-list nil
)
289 (defun mm-get-coding-system-list ()
290 "Get the coding system list."
291 (or mm-coding-system-list
292 (setq mm-coding-system-list
(mm-coding-system-list))))
294 (defun mm-coding-system-p (cs)
295 "Return non-nil if CS is a symbol naming a coding system.
296 In XEmacs, also return non-nil if CS is a coding system object.
297 If CS is available, return CS itself in Emacs, and return a coding
298 system object in XEmacs."
299 (if (fboundp 'find-coding-system
)
300 (and cs
(find-coding-system cs
))
301 (if (fboundp 'coding-system-p
)
302 (when (coding-system-p cs
)
305 (car (memq cs
(mm-get-coding-system-list))))))
307 (defun mm-codepage-setup (number &optional alias
)
308 "Create a coding system cpNUMBER.
309 The coding system is created using `codepage-setup'. If ALIAS is
310 non-nil, an alias is created and added to
311 `mm-charset-synonym-alist'. If ALIAS is a string, it's used as
312 the alias. Else windows-NUMBER is used."
314 (let ((completion-ignore-case t
)
315 (candidates (if (fboundp 'cp-supported-codepages
)
316 (cp-supported-codepages)
317 ;; Removed in Emacs 23 (unicode), so signal an error:
318 (error "`codepage-setup' not present in this Emacs version."))))
319 (list (completing-read "Setup DOS Codepage: (default 437) " candidates
320 nil t nil nil
"437"))))
322 (setq alias
(if (stringp alias
)
324 (intern (format "windows-%s" number
)))))
325 (let* ((cp (intern (format "cp%s" number
))))
326 (unless (mm-coding-system-p cp
)
327 (if (fboundp 'codepage-setup
) ; silence compiler
328 (codepage-setup number
)
329 (error "`codepage-setup' not present in this Emacs version.")))
331 ;; Don't add alias if setup of cp failed.
332 (mm-coding-system-p cp
))
333 (add-to-list 'mm-charset-synonym-alist
(cons alias cp
)))))
335 (defvar mm-charset-synonym-alist
337 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
338 ,@(unless (mm-coding-system-p 'x-ctext
)
339 '((x-ctext . ctext
)))
340 ;; ISO-8859-15 is very similar to ISO-8859-1. But it's _different_ in 8
342 ,@(unless (mm-coding-system-p 'iso-8859-15
)
343 '((iso-8859-15 . iso-8859-1
)))
344 ;; BIG-5HKSCS is similar to, but different than, BIG-5.
345 ,@(unless (mm-coding-system-p 'big5-hkscs
)
346 '((big5-hkscs . big5
)))
347 ;; A Microsoft misunderstanding.
348 ,@(when (and (not (mm-coding-system-p 'unicode
))
349 (mm-coding-system-p 'utf-16-le
))
350 '((unicode . utf-16-le
)))
351 ;; A Microsoft misunderstanding.
352 ,@(unless (mm-coding-system-p 'ks_c_5601-1987
)
353 (if (mm-coding-system-p 'cp949
)
354 '((ks_c_5601-1987 . cp949
))
355 '((ks_c_5601-1987 . euc-kr
))))
356 ;; Windows-31J is Windows Codepage 932.
357 ,@(when (and (not (mm-coding-system-p 'windows-31j
))
358 (mm-coding-system-p 'cp932
))
359 '((windows-31j . cp932
)))
360 ;; Charset name: GBK, Charset aliases: CP936, MS936, windows-936
361 ;; http://www.iana.org/assignments/charset-reg/GBK
362 ;; Emacs 22.1 has cp936, but not gbk, so we alias it:
363 ,@(when (and (not (mm-coding-system-p 'gbk
))
364 (mm-coding-system-p 'cp936
))
366 ;; UTF8 is a bogus name for UTF-8
367 ,@(when (and (not (mm-coding-system-p 'utf8
))
368 (mm-coding-system-p 'utf-8
))
370 ;; ISO8859-1 is a bogus name for ISO-8859-1
371 ,@(when (and (not (mm-coding-system-p 'iso8859-1
))
372 (mm-coding-system-p 'iso-8859-1
))
373 '((iso8859-1 . iso-8859-1
)))
374 ;; ISO_8859-1 is a bogus name for ISO-8859-1
375 ,@(when (and (not (mm-coding-system-p 'iso_8859-1
))
376 (mm-coding-system-p 'iso-8859-1
))
377 '((iso_8859-1 . iso-8859-1
)))
379 "A mapping from unknown or invalid charset names to the real charset names.
381 See `mm-codepage-iso-8859-list' and `mm-codepage-ibm-list'.")
383 (defcustom mm-codepage-iso-8859-list
384 (list 1250 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
385 ;; Outlook users in Czech republic. Use this to allow reading of
386 ;; their e-mails. cp1250 should be defined by M-x codepage-setup
388 '(1252 .
1) ;; Windows-1252 is a superset of iso-8859-1 (West
389 ;; Europe). See also `gnus-article-dumbquotes-map'.
390 '(1254 .
9) ;; Windows-1254 is a superset of iso-8859-9 (Turkish).
391 '(1255 .
8));; Windows-1255 is a superset of iso-8859-8 (Hebrew).
392 "A list of Windows codepage numbers and iso-8859 charset numbers.
394 If an element is a number corresponding to a supported windows
395 codepage, appropriate entries to `mm-charset-synonym-alist' are
396 added by `mm-setup-codepage-iso-8859'. An element may also be a
397 cons cell where the car is a codepage number and the cdr is the
398 corresponding number of an iso-8859 charset."
399 :type
'(list (set :inline t
400 (const 1250 :tag
"Central and East European")
401 (const (1252 .
1) :tag
"West European")
402 (const (1254 .
9) :tag
"Turkish")
403 (const (1255 .
8) :tag
"Hebrew"))
407 (integer :tag
"Windows codepage number")
408 (cons (integer :tag
"Windows codepage number")
409 (integer :tag
"iso-8859 charset number")))))
410 :version
"22.1" ;; Gnus 5.10.9
413 (defcustom mm-codepage-ibm-list
414 (list 437 ;; (US etc.)
418 863 ;; (Canadian French)
423 866 ;; (Cyrillic - Russian)
428 ;; In Emacs 23 (unicode), cp... and ibm... are aliases.
429 ;; Cf. http://thread.gmane.org/v9lkng5nwy.fsf@marauder.physik.uni-ulm.de
430 "List of IBM codepage numbers.
432 The codepage mappings slighly differ between IBM and other vendors.
433 See \"ftp://ftp.unicode.org/Public/MAPPINGS/VENDORS/IBM/README.TXT\".
435 If an element is a number corresponding to a supported windows
436 codepage, appropriate entries to `mm-charset-synonym-alist' are
437 added by `mm-setup-codepage-ibm'."
438 :type
'(list (set :inline t
439 (const 437 :tag
"US etc.")
440 (const 860 :tag
"Portugal")
441 (const 861 :tag
"Iceland")
442 (const 862 :tag
"Israel")
443 (const 863 :tag
"Canadian French")
444 (const 865 :tag
"Nordic")
446 (const 850 :tag
"Latin 1")
447 (const 855 :tag
"Cyrillic")
448 (const 866 :tag
"Cyrillic - Russian")
449 (const 857 :tag
"Turkish")
450 (const 864 :tag
"Arabic")
451 (const 869 :tag
"Greek")
452 (const 874 :tag
"Thai"))
455 (integer :tag
"Codepage number")))
456 :version
"22.1" ;; Gnus 5.10.9
459 (defun mm-setup-codepage-iso-8859 (&optional list
)
460 "Add appropriate entries to `mm-charset-synonym-alist'.
461 Unless LIST is given, `mm-codepage-iso-8859-list' is used."
463 (setq list mm-codepage-iso-8859-list
))
465 (let (cp windows iso
)
467 (setq cp
(intern (format "cp%d" (car i
)))
468 windows
(intern (format "windows-%d" (car i
)))
469 iso
(intern (format "iso-8859-%d" (cdr i
))))
470 (setq cp
(intern (format "cp%d" i
))
471 windows
(intern (format "windows-%d" i
))))
472 (unless (mm-coding-system-p windows
)
473 (if (mm-coding-system-p cp
)
474 (add-to-list 'mm-charset-synonym-alist
(cons windows cp
))
475 (add-to-list 'mm-charset-synonym-alist
(cons windows iso
)))))))
477 (defun mm-setup-codepage-ibm (&optional list
)
478 "Add appropriate entries to `mm-charset-synonym-alist'.
479 Unless LIST is given, `mm-codepage-ibm-list' is used."
481 (setq list mm-codepage-ibm-list
))
482 (dolist (number list
)
483 (let ((ibm (intern (format "ibm%d" number
)))
484 (cp (intern (format "cp%d" number
))))
485 (when (and (not (mm-coding-system-p ibm
))
486 (mm-coding-system-p cp
))
487 (add-to-list 'mm-charset-synonym-alist
(cons ibm cp
))))))
490 (mm-setup-codepage-iso-8859)
491 (mm-setup-codepage-ibm)
493 ;; Note: this has to be defined before `mm-charset-to-coding-system'.
494 (defcustom mm-charset-eval-alist
495 (if (featurep 'xemacs
)
496 nil
;; I don't know what would be useful for XEmacs.
497 '(;; Emacs 21 offers 1250 1251 1253 1257. Emacs 22 provides autoloads for
498 ;; 1250-1258 (i.e. `mm-codepage-setup' does nothing).
499 (windows-1250 .
(mm-codepage-setup 1250 t
))
500 (windows-1251 .
(mm-codepage-setup 1251 t
))
501 (windows-1253 .
(mm-codepage-setup 1253 t
))
502 (windows-1257 .
(mm-codepage-setup 1257 t
))))
503 "An alist of (CHARSET . FORM) pairs.
504 If an article is encoded in an unknown CHARSET, FORM is
505 evaluated. This allows to load additional libraries providing
506 charsets on demand. If supported by your Emacs version, you
507 could use `autoload-coding-system' here."
508 :version
"22.1" ;; Gnus 5.10.9
509 :type
'(list (set :inline t
510 (const (windows-1250 .
(mm-codepage-setup 1250 t
)))
511 (const (windows-1251 .
(mm-codepage-setup 1251 t
)))
512 (const (windows-1253 .
(mm-codepage-setup 1253 t
)))
513 (const (windows-1257 .
(mm-codepage-setup 1257 t
)))
514 (const (cp850 .
(mm-codepage-setup 850 nil
))))
517 (cons (symbol :tag
"charset")
518 (symbol :tag
"form"))))
520 (put 'mm-charset-eval-alist
'risky-local-variable t
)
522 (defvar mm-charset-override-alist
)
524 ;; Note: this function has to be defined before `mm-charset-override-alist'
525 ;; since it will use this function in order to determine its default value
526 ;; when loading mm-util.elc.
527 (defun mm-charset-to-coding-system (charset &optional lbt
528 allow-override silent
)
529 "Return coding-system corresponding to CHARSET.
530 CHARSET is a symbol naming a MIME charset.
531 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
532 used as the line break code type of the coding system.
534 If ALLOW-OVERRIDE is given, use `mm-charset-override-alist' to
535 map undesired charset names to their replacement. This should
536 only be used for decoding, not for encoding.
538 A non-nil value of SILENT means don't issue a warning even if CHARSET
540 ;; OVERRIDE is used (only) in `mm-decode-body' and `mm-decode-string'.
541 (when (stringp charset
)
542 (setq charset
(intern (downcase charset
))))
544 (setq charset
(intern (format "%s-%s" charset lbt
))))
548 ;; Running in a non-MULE environment.
549 ((or (null (mm-get-coding-system-list))
550 (not (fboundp 'coding-system-get
)))
552 ;; Check override list quite early. Should only used for decoding, not for
555 (let ((cs (cdr (assq charset mm-charset-override-alist
))))
556 (and cs
(mm-coding-system-p cs
) cs
))))
558 ((eq charset
'us-ascii
)
560 ;; Check to see whether we can handle this charset. (This depends
561 ;; on there being some coding system matching each `mime-charset'
562 ;; property defined, as there should be.)
563 ((and (mm-coding-system-p charset
)
564 ;;; Doing this would potentially weed out incorrect charsets.
566 ;;; (eq charset (coding-system-get charset 'mime-charset))
569 ;; Eval expressions from `mm-charset-eval-alist'
570 ((let* ((el (assq charset mm-charset-eval-alist
))
577 (condition-case nil
(eval form
) (error nil
))
578 ;; (message "Failed to eval `%s'" form))
579 (mm-coding-system-p cs
)
580 (message "Added charset `%s' via `mm-charset-eval-alist'" cs
))
582 ;; Translate invalid charsets.
583 ((let ((cs (cdr (assq charset mm-charset-synonym-alist
))))
585 (mm-coding-system-p cs
)
587 ;; "Using synonym `%s' from `mm-charset-synonym-alist' for `%s'"
590 ;; Last resort: search the coding system list for entries which
591 ;; have the right mime-charset in case the canonical name isn't
592 ;; defined (though it should be).
594 ;; mm-get-coding-system-list returns a list of cs without lbt.
596 (dolist (c (mm-get-coding-system-list))
598 (eq charset
(or (coding-system-get c
:mime-charset
)
599 (coding-system-get c
'mime-charset
))))
601 (unless (or silent cs
)
602 ;; Warn the user about unknown charset:
603 (if (fboundp 'gnus-message
)
604 (gnus-message 7 "Unknown charset: %s" charset
)
605 (message "Unknown charset: %s" charset
)))
608 ;; Note: `mm-charset-to-coding-system' has to be defined before this.
609 (defcustom mm-charset-override-alist
610 ;; Note: pairs that cannot be used in the Emacs version currently running
613 (iso-8859-1 . windows-1252
)
614 (iso-8859-8 . windows-1255
)
615 (iso-8859-9 . windows-1254
))
616 "A mapping from undesired charset names to their replacement.
618 You may add pairs like (iso-8859-1 . windows-1252) here,
619 i.e. treat iso-8859-1 as windows-1252. windows-1252 is a
620 superset of iso-8859-1."
627 (mapcar (lambda (pair)
628 (if (mm-charset-to-coding-system (cdr pair
)
632 (iso-8859-1 . windows-1252
)
633 (iso-8859-8 . windows-1255
)
634 (iso-8859-9 . windows-1254
)
635 (undecided . windows-1252
)))))
636 (val (copy-sequence (default-value 'mm-charset-override-alist
)))
639 (push (if (and (prog1
640 (setq pair
(assq (caar val
) defaults
))
641 (setq defaults
(delq pair defaults
)))
642 (equal (car val
) pair
))
645 (const :format
"(%v" ,(caar val
))
646 (symbol :size
3 :format
" . %v)\n" ,(cdar val
))))
648 (setq val
(cdr val
)))
650 (push `(const ,(pop defaults
)) rest
))
653 `(set :inline t
:format
"%v" ,@(nreverse rest
))
654 `(repeat :inline t
:tag
"Other options"
656 (symbol :size
3 :format
"(%v")
657 (symbol :size
3 :format
" . %v)\n")))))))
658 ;; Remove pairs that cannot be used in the Emacs version currently
659 ;; running. Note that this section will be evaluated when loading
661 :set
(lambda (symbol value
)
664 (mapcar (lambda (pair)
665 (if (mm-charset-to-coding-system (cdr pair
)
669 :version
"22.1" ;; Gnus 5.10.9
672 (defvar mm-binary-coding-system
674 ((mm-coding-system-p 'binary
) 'binary
)
675 ((mm-coding-system-p 'no-conversion
) 'no-conversion
)
677 "100% binary coding system.")
679 (defvar mm-text-coding-system
680 (or (if (memq system-type
'(windows-nt ms-dos ms-windows
))
681 (and (mm-coding-system-p 'raw-text-dos
) 'raw-text-dos
)
682 (and (mm-coding-system-p 'raw-text
) 'raw-text
))
683 mm-binary-coding-system
)
684 "Text-safe coding system (For removing ^M).")
686 (defvar mm-text-coding-system-for-write nil
687 "Text coding system for write.")
689 (defvar mm-auto-save-coding-system
691 ((mm-coding-system-p 'utf-8-emacs
) ; Mule 7
692 (if (memq system-type
'(windows-nt ms-dos ms-windows
))
693 (if (mm-coding-system-p 'utf-8-emacs-dos
)
694 'utf-8-emacs-dos mm-binary-coding-system
)
696 ((mm-coding-system-p 'emacs-mule
)
697 (if (memq system-type
'(windows-nt ms-dos ms-windows
))
698 (if (mm-coding-system-p 'emacs-mule-dos
)
699 'emacs-mule-dos mm-binary-coding-system
)
701 ((mm-coding-system-p 'escape-quoted
) 'escape-quoted
)
702 (t mm-binary-coding-system
))
703 "Coding system of auto save file.")
705 (defvar mm-universal-coding-system mm-auto-save-coding-system
706 "The universal coding system.")
708 ;; Fixme: some of the cars here aren't valid MIME charsets. That
709 ;; should only matter with XEmacs, though.
710 (defvar mm-mime-mule-charset-alist
712 (iso-8859-1 latin-iso8859-1
)
713 (iso-8859-2 latin-iso8859-2
)
714 (iso-8859-3 latin-iso8859-3
)
715 (iso-8859-4 latin-iso8859-4
)
716 (iso-8859-5 cyrillic-iso8859-5
)
717 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
718 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
719 ;; charset is koi8-r, not iso-8859-5.
720 (koi8-r cyrillic-iso8859-5 gnus-koi8-r
)
721 (iso-8859-6 arabic-iso8859-6
)
722 (iso-8859-7 greek-iso8859-7
)
723 (iso-8859-8 hebrew-iso8859-8
)
724 (iso-8859-9 latin-iso8859-9
)
725 (iso-8859-14 latin-iso8859-14
)
726 (iso-8859-15 latin-iso8859-15
)
727 (viscii vietnamese-viscii-lower
)
728 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978
)
729 (euc-kr korean-ksc5601
)
730 (gb2312 chinese-gb2312
)
732 (gb18030 gb18030-2-byte
733 gb18030-4-byte-bmp gb18030-4-byte-smp
734 gb18030-4-byte-ext-1 gb18030-4-byte-ext-2
)
735 (big5 chinese-big5-1 chinese-big5-2
)
737 (thai-tis620 thai-tis620
)
738 (windows-1251 cyrillic-iso8859-5
)
739 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column
)
740 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
741 latin-jisx0201 japanese-jisx0208-1978
742 chinese-gb2312 japanese-jisx0208
743 korean-ksc5601 japanese-jisx0212
)
744 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
745 latin-jisx0201 japanese-jisx0208-1978
746 chinese-gb2312 japanese-jisx0208
747 korean-ksc5601 japanese-jisx0212
748 chinese-cns11643-1 chinese-cns11643-2
)
749 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
750 cyrillic-iso8859-5 greek-iso8859-7
751 latin-jisx0201 japanese-jisx0208-1978
752 chinese-gb2312 japanese-jisx0208
753 korean-ksc5601 japanese-jisx0212
754 chinese-cns11643-1 chinese-cns11643-2
755 chinese-cns11643-3 chinese-cns11643-4
756 chinese-cns11643-5 chinese-cns11643-6
758 (iso-2022-jp-3 latin-jisx0201 japanese-jisx0208-1978 japanese-jisx0208
759 japanese-jisx0213-1 japanese-jisx0213-2
)
760 (shift_jis latin-jisx0201 katakana-jisx0201 japanese-jisx0208
)
761 ,(cond ((fboundp 'unicode-precedence-list
)
762 (cons 'utf-8
(delq 'ascii
(mapcar 'charset-name
763 (unicode-precedence-list)))))
764 ((or (not (fboundp 'charsetp
)) ;; non-Mule case
765 (charsetp 'unicode-a
)
766 (not (mm-coding-system-p 'mule-utf-8
)))
767 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e
))
768 (t ;; If we have utf-8 we're in Mule 5+.
771 (coding-system-get 'mule-utf-8
'safe-charsets
))))))
772 "Alist of MIME-charset/MULE-charsets.")
774 (defun mm-enrich-utf-8-by-mule-ucs ()
775 "Make the `utf-8' MIME charset usable by the Mule-UCS package.
776 This function will run when the `un-define' module is loaded under
777 XEmacs, and fill the `utf-8' entry in `mm-mime-mule-charset-alist'
778 with Mule charsets. It is completely useless for Emacs."
779 (when (boundp 'unicode-basic-translation-charset-order-list
)
785 'unicode-basic-translation-charset-order-list
))))
786 (elem (assq 'utf-8 mm-mime-mule-charset-alist
)))
789 (setq mm-mime-mule-charset-alist
790 (nconc mm-mime-mule-charset-alist
791 (list (cons 'utf-8 val
))))))
794 ;; Correct by construction, but should be unnecessary for Emacs:
795 (if (featurep 'xemacs
)
796 (eval-after-load "un-define" '(mm-enrich-utf-8-by-mule-ucs))
797 (when (and (fboundp 'coding-system-list
)
798 (fboundp 'sort-coding-systems
))
799 (let ((css (sort-coding-systems (coding-system-list 'base-only
)))
803 mime
(or (coding-system-get cs
:mime-charset
); Emacs 23 (unicode)
804 (coding-system-get cs
'mime-charset
)))
806 (not (eq t
(setq mule
807 (coding-system-get cs
'safe-charsets
))))
808 (not (assq mime alist
)))
809 (push (cons mime
(delq 'ascii mule
)) alist
)))
810 (setq mm-mime-mule-charset-alist
(nreverse alist
)))))
812 (defvar mm-hack-charsets
'(iso-8859-15 iso-2022-jp-2
)
813 "A list of special charsets.
814 Valid elements include:
815 `iso-8859-15' convert ISO-8859-1, -9 to ISO-8859-15 if ISO-8859-15 exists.
816 `iso-2022-jp-2' convert ISO-2022-jp to ISO-2022-jp-2 if ISO-2022-jp-2 exists."
819 (defvar mm-iso-8859-15-compatible
820 '((iso-8859-1 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE")
821 (iso-8859-9 "\xA4\xA6\xA8\xB4\xB8\xBC\xBD\xBE\xD0\xDD\xDE\xF0\xFD\xFE"))
822 "ISO-8859-15 exchangeable coding systems and inconvertible characters.")
824 (defvar mm-iso-8859-x-to-15-table
825 (and (fboundp 'coding-system-p
)
826 (mm-coding-system-p 'iso-8859-15
)
829 (if (mm-coding-system-p (car cs
))
830 (let ((c (string-to-char
831 (decode-coding-string "\341" (car cs
)))))
832 (cons (char-charset c
)
835 (decode-coding-string "\341" 'iso-8859-15
)) c
)
836 (string-to-list (decode-coding-string (car (cdr cs
))
839 mm-iso-8859-15-compatible
))
840 "A table of the difference character between ISO-8859-X and ISO-8859-15.")
842 (defcustom mm-coding-system-priorities
843 (if (boundp 'current-language-environment
)
844 (let ((lang (symbol-value 'current-language-environment
)))
845 (cond ((string= lang
"Japanese")
846 ;; Japanese users prefer iso-2022-jp to euc-japan or
847 ;; shift_jis, however iso-8859-1 should be used when
848 ;; there are only ASCII text and Latin-1 characters.
849 '(iso-8859-1 iso-2022-jp iso-2022-jp-2 shift_jis utf-8
)))))
850 "Preferred coding systems for encoding outgoing messages.
852 More than one suitable coding system may be found for some text.
853 By default, the coding system with the highest priority is used
854 to encode outgoing messages (see `sort-coding-systems'). If this
855 variable is set, it overrides the default priority."
857 :type
'(repeat (symbol :tag
"Coding system"))
861 (defvar mm-use-find-coding-systems-region
862 (fboundp 'find-coding-systems-region
)
863 "Use `find-coding-systems-region' to find proper coding systems.
865 Setting it to nil is useful on Emacsen supporting Unicode if sending
866 mail with multiple parts is preferred to sending a Unicode one.")
868 ;;; Internal variables:
872 (defun mm-mule-charset-to-mime-charset (charset)
873 "Return the MIME charset corresponding to the given Mule CHARSET."
874 (if (and (fboundp 'find-coding-systems-for-charsets
)
875 (fboundp 'sort-coding-systems
))
876 (let ((css (sort (sort-coding-systems
877 (find-coding-systems-for-charsets (list charset
)))
878 'mm-sort-coding-systems-predicate
))
880 (while (and (not mime
)
882 (when (setq cs
(pop css
))
883 (setq mime
(or (coding-system-get cs
:mime-charset
)
884 (coding-system-get cs
'mime-charset
)))))
886 (let ((alist (mapcar (lambda (cs)
887 (assq cs mm-mime-mule-charset-alist
))
888 (sort (mapcar 'car mm-mime-mule-charset-alist
)
889 'mm-sort-coding-systems-predicate
)))
892 (when (memq charset
(cdar alist
))
893 (setq out
(caar alist
)
899 (defvar mm-emacs-mule
(and (not (featurep 'xemacs
))
900 (boundp 'default-enable-multibyte-characters
)
901 default-enable-multibyte-characters
902 (fboundp 'set-buffer-multibyte
))
903 "True in Emacs with Mule.")
906 (defun mm-enable-multibyte ()
907 "Set the multibyte flag of the current buffer.
908 Only do this if the default value of `enable-multibyte-characters' is
909 non-nil. This is a no-op in XEmacs."
910 (set-buffer-multibyte 'to
))
911 (defalias 'mm-enable-multibyte
'ignore
))
914 (defun mm-disable-multibyte ()
915 "Unset the multibyte flag of in the current buffer.
916 This is a no-op in XEmacs."
917 (set-buffer-multibyte nil
))
918 (defalias 'mm-disable-multibyte
'ignore
)))
920 (defun mm-preferred-coding-system (charset)
921 ;; A typo in some Emacs versions.
922 (or (get-charset-property charset
'preferred-coding-system
)
923 (get-charset-property charset
'prefered-coding-system
)))
925 ;; Mule charsets shouldn't be used.
926 (defsubst mm-guess-charset
()
927 "Guess Mule charset from the language environment."
929 mail-parse-mule-charset
;; cached mule-charset
931 (setq mail-parse-mule-charset
932 (and (boundp 'current-language-environment
)
935 (assoc current-language-environment
936 language-info-alist
))))))
937 (if (or (not mail-parse-mule-charset
)
938 (eq mail-parse-mule-charset
'ascii
))
939 (setq mail-parse-mule-charset
940 (or (car (last (assq mail-parse-charset
941 mm-mime-mule-charset-alist
)))
944 mail-parse-mule-charset
)))
946 (defun mm-charset-after (&optional pos
)
947 "Return charset of a character in current buffer at position POS.
948 If POS is nil, it defauls to the current point.
949 If POS is out of range, the value is nil.
950 If the charset is `composition', return the actual one."
951 (let ((char (char-after pos
)) charset
)
952 (if (< (mm-char-int char
) 128)
953 (setq charset
'ascii
)
954 ;; charset-after is fake in some Emacsen.
955 (setq charset
(and (fboundp 'char-charset
) (char-charset char
)))
956 (if (eq charset
'composition
) ; Mule 4
957 (let ((p (or pos
(point))))
958 (cadr (find-charset-region p
(1+ p
))))
959 (if (and charset
(not (memq charset
'(ascii eight-bit-control
960 eight-bit-graphic
))))
962 (mm-guess-charset))))))
964 (defun mm-mime-charset (charset)
965 "Return the MIME charset corresponding to the given Mule CHARSET."
966 (if (eq charset
'unknown
)
967 (error "The message contains non-printable characters, please use attachment"))
968 (if (and (fboundp 'coding-system-get
) (fboundp 'get-charset-property
))
969 ;; This exists in Emacs 20.
971 (and (mm-preferred-coding-system charset
)
972 (or (coding-system-get
973 (mm-preferred-coding-system charset
) :mime-charset
)
975 (mm-preferred-coding-system charset
) 'mime-charset
)))
976 (and (eq charset
'ascii
)
978 (mm-preferred-coding-system charset
)
979 (mm-mule-charset-to-mime-charset charset
))
980 ;; This is for XEmacs.
981 (mm-mule-charset-to-mime-charset charset
)))
983 (if (fboundp 'delete-dups
)
984 (defalias 'mm-delete-duplicates
'delete-dups
)
985 (defun mm-delete-duplicates (list)
986 "Destructively remove `equal' duplicates from LIST.
987 Store the result in LIST and return it. LIST must be a proper list.
988 Of several `equal' occurrences of an element in LIST, the first
991 This is a compatibility function for Emacsen without `delete-dups'."
992 ;; Code from `subr.el' in Emacs 22:
995 (setcdr tail
(delete (car tail
) (cdr tail
)))
996 (setq tail
(cdr tail
))))
999 ;; Fixme: This is used in places when it should be testing the
1000 ;; default multibyteness. See mm-default-multibyte-p.
1002 (if (and (not (featurep 'xemacs
))
1003 (boundp 'enable-multibyte-characters
))
1004 (defun mm-multibyte-p ()
1005 "Non-nil if multibyte is enabled in the current buffer."
1006 enable-multibyte-characters
)
1007 (defun mm-multibyte-p () (featurep 'mule
))))
1009 (defun mm-default-multibyte-p ()
1010 "Return non-nil if the session is multibyte.
1011 This affects whether coding conversion should be attempted generally."
1012 (if (featurep 'mule
)
1013 (if (boundp 'default-enable-multibyte-characters
)
1014 default-enable-multibyte-characters
1017 (defun mm-iso-8859-x-to-15-region (&optional b e
)
1018 (if (fboundp 'char-charset
)
1019 (let (charset item c inconvertible
)
1021 (if e
(narrow-to-region b e
))
1022 (goto-char (point-min))
1023 (skip-chars-forward "\0-\177")
1026 ((not (setq item
(assq (char-charset (setq c
(char-after)))
1027 mm-iso-8859-x-to-15-table
)))
1029 ((memq c
(cdr (cdr item
)))
1030 (setq inconvertible t
)
1033 (insert-before-markers (prog1 (+ c
(car (cdr item
)))
1035 (skip-chars-forward "\0-\177")))
1036 (not inconvertible
))))
1038 (defun mm-sort-coding-systems-predicate (a b
)
1040 (mapcar (lambda (cs)
1041 ;; Note: invalid entries are dropped silently
1042 (and (setq cs
(mm-coding-system-p cs
))
1043 (coding-system-base cs
)))
1044 mm-coding-system-priorities
)))
1045 (and (setq a
(mm-coding-system-p a
))
1046 (if (setq b
(mm-coding-system-p b
))
1047 (> (length (memq (coding-system-base a
) priorities
))
1048 (length (memq (coding-system-base b
) priorities
)))
1052 (autoload 'latin-unity-massage-name
"latin-unity")
1053 (autoload 'latin-unity-maybe-remap
"latin-unity")
1054 (autoload 'latin-unity-representations-feasible-region
"latin-unity")
1055 (autoload 'latin-unity-representations-present-region
"latin-unity"))
1057 (defvar latin-unity-coding-systems
)
1058 (defvar latin-unity-ucs-list
)
1060 (defun mm-xemacs-find-mime-charset-1 (begin end
)
1061 "Determine which MIME charset to use to send region as message.
1062 This uses the XEmacs-specific latin-unity package to better handle the
1063 case where identical characters from diverse ISO-8859-? character sets
1064 can be encoded using a single one of the corresponding coding systems.
1066 It treats `mm-coding-system-priorities' as the list of preferred
1067 coding systems; a useful example setting for this list in Western
1068 Europe would be '(iso-8859-1 iso-8859-15 utf-8), which would default
1069 to the very standard Latin 1 coding system, and only move to coding
1070 systems that are less supported as is necessary to encode the
1071 characters that exist in the buffer.
1073 Latin Unity doesn't know about those non-ASCII Roman characters that
1074 are available in various East Asian character sets. As such, its
1075 behavior if you have a JIS 0212 LATIN SMALL LETTER A WITH ACUTE in a
1076 buffer and it can otherwise be encoded as Latin 1, won't be ideal.
1077 But this is very much a corner case, so don't worry about it."
1078 (let ((systems mm-coding-system-priorities
) csets psets curset
)
1080 ;; Load the Latin Unity library, if available.
1081 (when (and (not (featurep 'latin-unity
)) (locate-library "latin-unity"))
1082 (require 'latin-unity
))
1084 ;; Now, can we use it?
1085 (if (featurep 'latin-unity
)
1087 (setq csets
(latin-unity-representations-feasible-region begin end
)
1088 psets
(latin-unity-representations-present-region begin end
))
1092 ;; Pass back the first coding system in the preferred list
1093 ;; that can encode the whole region.
1094 (dolist (curset systems
)
1095 (setq curset
(latin-unity-massage-name 'buffer-default curset
))
1097 ;; If the coding system is a universal coding system, then
1098 ;; it can certainly encode all the characters in the region.
1099 (if (memq curset latin-unity-ucs-list
)
1100 (throw 'done
(list curset
)))
1102 ;; If a coding system isn't universal, and isn't in
1103 ;; the list that latin unity knows about, we can't
1104 ;; decide whether to use it here. Leave that until later
1105 ;; in `mm-find-mime-charset-region' function, whence we
1106 ;; have been called.
1107 (unless (memq curset latin-unity-coding-systems
)
1110 ;; Right, we know about this coding system, and it may
1111 ;; conceivably be able to encode all the characters in
1113 (if (latin-unity-maybe-remap begin end curset csets psets t
)
1114 (throw 'done
(list curset
))))
1116 ;; Can't encode using anything from the
1117 ;; `mm-coding-system-priorities' list.
1118 ;; Leave `mm-find-mime-charset' to do most of the work.
1121 ;; Right, latin unity isn't available; let `mm-find-charset-region'
1122 ;; take its default action, which equally applies to GNU Emacs.
1125 (defmacro mm-xemacs-find-mime-charset
(begin end
)
1126 (when (featurep 'xemacs
)
1127 `(and (featurep 'mule
) (mm-xemacs-find-mime-charset-1 ,begin
,end
))))
1129 (declare-function mm-delete-duplicates
"mm-util" (list))
1131 (defun mm-find-mime-charset-region (b e
&optional hack-charsets
)
1132 "Return the MIME charsets needed to encode the region between B and E.
1133 nil means ASCII, a single-element list represents an appropriate MIME
1134 charset, and a longer list means no appropriate charset."
1136 ;; The return possibilities of this function are a mess...
1137 (or (and (mm-multibyte-p)
1138 mm-use-find-coding-systems-region
1139 ;; Find the mime-charset of the most preferred coding
1140 ;; system that has one.
1141 (let ((systems (find-coding-systems-region b e
)))
1142 (when mm-coding-system-priorities
1144 (sort systems
'mm-sort-coding-systems-predicate
)))
1145 (setq systems
(delq 'compound-text systems
))
1146 (unless (equal systems
'(undecided))
1148 (let* ((head (pop systems
))
1149 (cs (or (coding-system-get head
:mime-charset
)
1150 (coding-system-get head
'mime-charset
))))
1151 ;; The mime-charset (`x-ctext') of
1152 ;; `compound-text' is not in the IANA list. We
1153 ;; shouldn't normally use anything here with a
1154 ;; mime-charset having an `x-' prefix.
1155 ;; Fixme: Allow this to be overridden, since
1156 ;; there is existing use of x-ctext.
1157 ;; Also people apparently need the coding system
1158 ;; `iso-2022-jp-3' (which Mule-UCS defines with
1159 ;; mime-charset, though it's not valid).
1161 (not (string-match "^[Xx]-" (symbol-name cs
)))
1162 ;; UTF-16 of any variety is invalid for
1163 ;; text parts and, unfortunately, has
1164 ;; mime-charset defined both in Mule-UCS
1165 ;; and versions of Emacs. (The name
1166 ;; might be `mule-utf-16...' or
1168 (not (string-match "utf-16" (symbol-name cs
))))
1170 charsets
(list cs
))))))
1172 ;; If we're XEmacs, and some coding system is appropriate,
1173 ;; mm-xemacs-find-mime-charset will return an appropriate list.
1174 ;; Otherwise, we'll get nil, and the next setq will get invoked.
1175 (setq charsets
(mm-xemacs-find-mime-charset b e
))
1177 ;; Fixme: won't work for unibyte Emacs 23:
1179 ;; We're not multibyte, or a single coding system won't cover it.
1181 (mm-delete-duplicates
1182 (mapcar 'mm-mime-charset
1184 (mm-find-charset-region b e
))))))
1185 (if (and (> (length charsets
) 1)
1186 (memq 'iso-8859-15 charsets
)
1187 (memq 'iso-8859-15 hack-charsets
)
1188 (save-excursion (mm-iso-8859-x-to-15-region b e
)))
1189 (dolist (x mm-iso-8859-15-compatible
)
1190 (setq charsets
(delq (car x
) charsets
))))
1191 (if (and (memq 'iso-2022-jp-2 charsets
)
1192 (memq 'iso-2022-jp-2 hack-charsets
))
1193 (setq charsets
(delq 'iso-2022-jp charsets
)))
1194 ;; Attempt to reduce the number of charsets if utf-8 is available.
1195 (if (and (featurep 'xemacs
)
1196 (> (length charsets
) 1)
1197 (mm-coding-system-p 'utf-8
))
1198 (let ((mm-coding-system-priorities
1199 (cons 'utf-8 mm-coding-system-priorities
)))
1201 (mm-delete-duplicates
1202 (mapcar 'mm-mime-charset
1204 (mm-find-charset-region b e
)))))))
1207 (defmacro mm-with-unibyte-buffer
(&rest forms
)
1208 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1209 Use unibyte mode for this."
1211 (mm-disable-multibyte)
1213 (put 'mm-with-unibyte-buffer
'lisp-indent-function
0)
1214 (put 'mm-with-unibyte-buffer
'edebug-form-spec
'(body))
1216 (defmacro mm-with-multibyte-buffer
(&rest forms
)
1217 "Create a temporary buffer, and evaluate FORMS there like `progn'.
1218 Use multibyte mode for this."
1220 (mm-enable-multibyte)
1222 (put 'mm-with-multibyte-buffer
'lisp-indent-function
0)
1223 (put 'mm-with-multibyte-buffer
'edebug-form-spec
'(body))
1225 (defmacro mm-with-unibyte-current-buffer
(&rest forms
)
1226 "Evaluate FORMS with current buffer temporarily made unibyte.
1227 Also bind `default-enable-multibyte-characters' to nil.
1228 Equivalent to `progn' in XEmacs
1230 NOTE: Use this macro with caution in multibyte buffers (it is not
1231 worth using this macro in unibyte buffers of course). Use of
1232 `(set-buffer-multibyte t)', which is run finally, is generally
1233 harmful since it is likely to modify existing data in the buffer.
1234 For instance, it converts \"\\300\\255\" into \"\\255\" in
1235 Emacs 23 (unicode)."
1236 (let ((multibyte (make-symbol "multibyte"))
1237 (buffer (make-symbol "buffer")))
1239 (let ((,multibyte enable-multibyte-characters
)
1240 (,buffer
(current-buffer)))
1242 (let (default-enable-multibyte-characters)
1243 (set-buffer-multibyte nil
)
1245 (set-buffer ,buffer
)
1246 (set-buffer-multibyte ,multibyte
)))
1247 (let (default-enable-multibyte-characters)
1249 (put 'mm-with-unibyte-current-buffer
'lisp-indent-function
0)
1250 (put 'mm-with-unibyte-current-buffer
'edebug-form-spec
'(body))
1252 (defun mm-find-charset-region (b e
)
1253 "Return a list of Emacs charsets in the region B to E."
1255 ((and (mm-multibyte-p)
1256 (fboundp 'find-charset-region
))
1257 ;; Remove composition since the base charsets have been included.
1258 ;; Remove eight-bit-*, treat them as ascii.
1259 (let ((css (find-charset-region b e
)))
1261 '(composition eight-bit-control eight-bit-graphic control-1
)
1263 (setq css
(delq cs css
)))))
1265 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
1268 (narrow-to-region b e
)
1269 (goto-char (point-min))
1270 (skip-chars-forward "\0-\177")
1275 (and (boundp 'current-language-environment
)
1276 (car (last (assq 'charset
1277 (assoc current-language-environment
1278 language-info-alist
))))))
1279 (if (eq charset
'ascii
) (setq charset nil
))
1282 (car (last (assq mail-parse-charset
1283 mm-mime-mule-charset-alist
)))))
1284 (list 'ascii
(or charset
'latin-iso8859-1
)))))))))
1286 (defun mm-auto-mode-alist ()
1287 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
1288 (let ((alist auto-mode-alist
)
1291 (when (listp (cdar alist
))
1292 (push (car alist
) out
))
1296 (defvar mm-inhibit-file-name-handlers
1297 '(jka-compr-handler image-file-handler epa-file-handler
)
1298 "A list of handlers doing (un)compression (etc) thingies.")
1300 (defun mm-insert-file-contents (filename &optional visit beg end replace
1302 "Like `insert-file-contents', but only reads in the file.
1303 A buffer may be modified in several ways after reading into the buffer due
1304 to advanced Emacs features, such as file-name-handlers, format decoding,
1305 `find-file-hooks', etc.
1306 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
1307 This function ensures that none of these modifications will take place."
1308 (let* ((format-alist nil
)
1309 (auto-mode-alist (if inhibit nil
(mm-auto-mode-alist)))
1310 (default-major-mode 'fundamental-mode
)
1311 (enable-local-variables nil
)
1312 (after-insert-file-functions nil
)
1313 (enable-local-eval nil
)
1314 (inhibit-file-name-operation (if inhibit
1315 'insert-file-contents
1316 inhibit-file-name-operation
))
1317 (inhibit-file-name-handlers
1319 (append mm-inhibit-file-name-handlers
1320 inhibit-file-name-handlers
)
1321 inhibit-file-name-handlers
))
1322 (ffh (if (boundp 'find-file-hook
)
1325 (val (symbol-value ffh
)))
1328 (insert-file-contents filename visit beg end replace
)
1331 (defun mm-append-to-file (start end filename
&optional codesys inhibit
)
1332 "Append the contents of the region to the end of file FILENAME.
1333 When called from a function, expects three arguments,
1334 START, END and FILENAME. START and END are buffer positions
1335 saying what text to write.
1336 Optional fourth argument specifies the coding system to use when
1338 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1339 (let ((coding-system-for-write
1340 (or codesys mm-text-coding-system-for-write
1341 mm-text-coding-system
))
1342 (inhibit-file-name-operation (if inhibit
1344 inhibit-file-name-operation
))
1345 (inhibit-file-name-handlers
1347 (append mm-inhibit-file-name-handlers
1348 inhibit-file-name-handlers
)
1349 inhibit-file-name-handlers
)))
1350 (write-region start end filename t
'no-message
)
1351 (message "Appended to %s" filename
)))
1353 (defun mm-write-region (start end filename
&optional append visit lockname
1354 coding-system inhibit
)
1356 "Like `write-region'.
1357 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'."
1358 (let ((coding-system-for-write
1359 (or coding-system mm-text-coding-system-for-write
1360 mm-text-coding-system
))
1361 (inhibit-file-name-operation (if inhibit
1363 inhibit-file-name-operation
))
1364 (inhibit-file-name-handlers
1366 (append mm-inhibit-file-name-handlers
1367 inhibit-file-name-handlers
)
1368 inhibit-file-name-handlers
)))
1369 (write-region start end filename append visit lockname
)))
1371 (autoload 'gmm-write-region
"gmm-utils")
1373 ;; It is not a MIME function, but some MIME functions use it.
1374 (if (and (fboundp 'make-temp-file
)
1376 (let ((def (symbol-function 'make-temp-file
)))
1377 (and (byte-code-function-p def
)
1378 (setq def
(if (fboundp 'compiled-function-arglist
)
1380 (eval (list 'compiled-function-arglist def
))
1383 (eq (nth 3 def
) 'suffix
)))))
1384 (defalias 'mm-make-temp-file
'make-temp-file
)
1385 ;; Stolen (and modified for XEmacs) from Emacs 22.
1386 (defun mm-make-temp-file (prefix &optional dir-flag suffix
)
1387 "Create a temporary file.
1388 The returned file name (created by appending some random characters at the end
1389 of PREFIX, and expanding against `temporary-file-directory' if necessary),
1390 is guaranteed to point to a newly created empty file.
1391 You can then use `write-region' to write new data into the file.
1393 If DIR-FLAG is non-nil, create a new empty directory instead of a file.
1395 If SUFFIX is non-nil, add that at the end of the file name."
1396 (let ((umask (default-file-modes))
1400 ;; Create temp files with strict access rights. It's easy to
1401 ;; loosen them later, whereas it's impossible to close the
1402 ;; time-window of loose permissions otherwise.
1403 (set-default-file-modes 448)
1404 (while (condition-case err
1410 (if (fboundp 'temp-directory
)
1413 temporary-file-directory
))))
1415 (setq file
(concat file suffix
)))
1417 (make-directory file
)
1418 ;; NOTE: This is unsafe if Emacs 20
1419 ;; users and XEmacs users don't use
1420 ;; a secure temp directory.
1421 (gmm-write-region "" nil file nil
'silent
1424 (file-already-exists t
)
1425 ;; The XEmacs version of `make-directory' issues
1427 (file-error (or (and (featurep 'xemacs
)
1428 (file-exists-p file
))
1429 (signal (car err
) (cdr err
)))))
1430 ;; the file was somehow created by someone else between
1431 ;; `make-temp-name' and `write-region', let's try again.
1435 (set-default-file-modes umask
)))))
1437 (defun mm-image-load-path (&optional package
)
1439 (dolist (path load-path
(nreverse result
))
1442 (setq dir
(concat (file-name-directory
1443 (directory-file-name path
))
1444 "etc/images/" (or package
"gnus/")))))
1446 (push path result
))))
1448 ;; Fixme: This doesn't look useful where it's used.
1449 (if (fboundp 'detect-coding-region
)
1450 (defun mm-detect-coding-region (start end
)
1451 "Like `detect-coding-region' except returning the best one."
1452 (let ((coding-systems
1453 (detect-coding-region start end
)))
1454 (or (car-safe coding-systems
)
1456 (defun mm-detect-coding-region (start end
)
1457 (let ((point (point)))
1459 (skip-chars-forward "\0-\177" end
)
1461 (if (eq (point) end
) 'ascii
(mm-guess-charset))
1462 (goto-char point
)))))
1464 (declare-function mm-detect-coding-region
"mm-util" (start end
))
1466 (if (fboundp 'coding-system-get
)
1467 (defun mm-detect-mime-charset-region (start end
)
1468 "Detect MIME charset of the text in the region between START and END."
1469 (let ((cs (mm-detect-coding-region start end
)))
1470 (or (coding-system-get cs
:mime-charset
)
1471 (coding-system-get cs
'mime-charset
))))
1472 (defun mm-detect-mime-charset-region (start end
)
1473 "Detect MIME charset of the text in the region between START and END."
1474 (let ((cs (mm-detect-coding-region start end
)))
1478 (unless (fboundp 'coding-system-to-mime-charset
)
1479 (defalias 'coding-system-to-mime-charset
'ignore
)))
1481 (defun mm-coding-system-to-mime-charset (coding-system)
1482 "Return the MIME charset corresponding to CODING-SYSTEM.
1483 To make this function work with XEmacs, the APEL package is required."
1485 (or (and (fboundp 'coding-system-get
)
1486 (or (coding-system-get coding-system
:mime-charset
)
1487 (coding-system-get coding-system
'mime-charset
)))
1488 (and (featurep 'xemacs
)
1489 (or (and (fboundp 'coding-system-to-mime-charset
)
1490 (not (eq (symbol-function 'coding-system-to-mime-charset
)
1492 (and (condition-case nil
1495 (fboundp 'coding-system-to-mime-charset
)))
1496 (coding-system-to-mime-charset coding-system
)))))
1499 (require 'jka-compr
))
1501 (defun mm-decompress-buffer (filename &optional inplace force
)
1502 "Decompress buffer's contents, depending on jka-compr.
1503 Only when FORCE is t or `auto-compression-mode' is enabled and FILENAME
1504 agrees with `jka-compr-compression-info-list', decompression is done.
1505 Signal an error if FORCE is neither nil nor t and compressed data are
1506 not decompressed because `auto-compression-mode' is disabled.
1507 If INPLACE is nil, return decompressed data or nil without modifying
1508 the buffer. Otherwise, replace the buffer's contents with the
1509 decompressed data. The buffer's multibyteness must be turned off."
1512 (prog1 t
(require 'jka-compr
))
1513 (and (fboundp 'jka-compr-installed-p
)
1514 (jka-compr-installed-p))))
1515 (let ((info (jka-compr-get-compression-info filename
)))
1517 (unless (or (memq force
(list nil t
))
1518 (jka-compr-installed-p))
1520 (let ((prog (jka-compr-info-uncompress-program info
))
1521 (args (jka-compr-info-uncompress-args info
))
1522 (msg (format "%s %s..."
1523 (jka-compr-info-uncompress-message info
)
1525 (err-file (jka-compr-make-temp-name))
1526 (cur (current-buffer))
1527 (coding-system-for-read mm-binary-coding-system
)
1528 (coding-system-for-write mm-binary-coding-system
)
1531 (mm-with-unibyte-buffer
1532 (insert-buffer-substring cur
)
1535 (unless (memq (apply 'call-process-region
1536 (point-min) (point-max)
1537 prog t
(list t err-file
) nil args
)
1538 jka-compr-acceptable-retval-list
)
1542 (delete "" (split-string
1544 (insert-file-contents err-file
)
1550 (format "Error while executing \"%s %s < %s\""
1551 prog
(mapconcat 'identity args
" ")
1553 (setq retval
(buffer-string)))
1555 (setq err-msg
(error-message-string err
)))))
1556 (when (file-exists-p err-file
)
1557 (ignore-errors (jka-compr-delete-temp-file err-file
)))
1560 (delete-region (point-min) (point-max))
1563 (message "%s" (or err-msg
(concat msg
"done")))
1567 (unless (fboundp 'coding-system-name
)
1568 (defalias 'coding-system-name
'ignore
))
1569 (unless (fboundp 'find-file-coding-system-for-read-from-filename
)
1570 (defalias 'find-file-coding-system-for-read-from-filename
'ignore
))
1571 (unless (fboundp 'find-operation-coding-system
)
1572 (defalias 'find-operation-coding-system
'ignore
)))
1574 (defun mm-find-buffer-file-coding-system (&optional filename
)
1575 "Find coding system used to decode the contents of the current buffer.
1576 This function looks for the coding system magic cookie or examines the
1577 coding system specified by `file-coding-system-alist' being associated
1578 with FILENAME which defaults to `buffer-file-name'. Data compressed by
1579 gzip, bzip2, etc. are allowed."
1581 (setq filename buffer-file-name
))
1583 (let ((decomp (unless ;; No worth to examine charset of tar files.
1586 "\\.\\(?:tar\\.[^.]+\\|tbz\\|tgz\\)\\'"
1588 (mm-decompress-buffer filename nil t
))))
1590 (set-buffer (let (default-enable-multibyte-characters)
1591 (generate-new-buffer " *temp*")))
1593 (setq filename
(file-name-sans-extension filename
)))
1594 (goto-char (point-min))
1597 ((boundp 'set-auto-coding-function
) ;; Emacs
1599 (or (funcall (symbol-value 'set-auto-coding-function
)
1600 filename
(- (point-max) (point-min)))
1601 (car (find-operation-coding-system 'insert-file-contents
1603 (let (auto-coding-alist)
1605 (funcall (symbol-value 'set-auto-coding-function
)
1606 nil
(- (point-max) (point-min)))
1608 ((and (featurep 'xemacs
) (featurep 'file-coding
)) ;; XEmacs
1609 (let ((case-fold-search t
)
1610 (end (point-at-eol))
1613 (and (re-search-forward "-\\*-+[\t ]*" end t
)
1615 (setq start
(match-end 0))
1616 (re-search-forward "[\t ]*-+\\*-" end t
))
1618 (setq end
(match-beginning 0))
1620 (or (looking-at "coding:[\t ]*\\([^\t ;]+\\)")
1622 "[\t ;]+coding:[\t ]*\\([^\t ;]+\\)"
1624 (find-coding-system (setq codesys
1625 (intern (match-string 1))))
1627 (and (re-search-forward "^[\t ]*;+[\t ]*Local[\t ]+Variables:"
1630 (setq start
(match-end 0))
1631 (re-search-forward "^[\t ]*;+[\t ]*End:" nil t
))
1633 (setq end
(match-beginning 0))
1636 "^[\t ]*;+[\t ]*coding:[\t ]*\\([^\t\n\r ]+\\)"
1638 (find-coding-system (setq codesys
1639 (intern (match-string 1))))
1642 (goto-char (point-min))
1643 (setq case-fold-search nil
)
1644 (re-search-forward "^;;;coding system: "
1645 ;;(+ (point-min) 3000) t))
1647 (looking-at "[^\t\n\r ]+")
1649 (setq codesys
(intern (match-string 0))))
1653 (find-file-coding-system-for-read-from-filename
1655 (coding-system-name (coding-system-base codesys
)))))))
1657 (kill-buffer (current-buffer)))))))
1661 ;; arch-tag: 94dc5388-825d-4fd1-bfa5-2100aa351238
1662 ;;; mm-util.el ends here