1 ;;; mm-util.el --- Utility functions for Mule and low level things
2 ;; Copyright (C) 1998, 1999, 2000, 2001, 2002 Free Software Foundation, Inc.
4 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
5 ;; MORIOKA Tomohiko <morioka@jaist.ac.jp>
6 ;; This file is part of GNU Emacs.
8 ;; GNU Emacs is free software; you can redistribute it and/or modify
9 ;; it under the terms of the GNU General Public License as published by
10 ;; the Free Software Foundation; either version 2, or (at your option)
13 ;; GNU Emacs is distributed in the hope that it will be useful,
14 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
15 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 ;; GNU General Public License for more details.
18 ;; You should have received a copy of the GNU General Public License
19 ;; along with GNU Emacs; see the file COPYING. If not, write to the
20 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
21 ;; Boston, MA 02111-1307, USA.
29 (defvar mm-mime-mule-charset-alist))
35 (let ((nfunc (intern (format "mm-%s" (car elem)))))
36 (if (fboundp (car elem))
37 (defalias nfunc (car elem))
38 (defalias nfunc (cdr elem)))))
39 '((decode-coding-string . (lambda (s a) s))
40 (encode-coding-string . (lambda (s a) s))
41 (encode-coding-region . ignore)
42 (coding-system-list . ignore)
43 (decode-coding-region . ignore)
45 (device-type . ignore)
46 (coding-system-equal . equal)
47 (annotationp . ignore)
48 (set-buffer-file-coding-system . ignore)
50 . (lambda (charset int)
58 (mapcar (lambda (e) (list (symbol-name (car e))))
59 mm-mime-mule-charset-alist)
62 . (lambda (from to string) ;; stolen (and renamed) from nnheader.el
63 "Replace characters in STRING from FROM to TO."
64 (let ((string (substring string 0)) ;Copy string.
67 ;; Replace all occurrences of FROM with TO.
69 (when (= (aref string idx) from)
73 (string-as-unibyte . identity)
74 (string-as-multibyte . identity)
75 (multibyte-string-p . ignore)
76 (insert-byte . insert))))
79 (defalias 'mm-char-or-char-int-p
81 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
82 ((fboundp 'char-valid-p) 'char-valid-p)
86 (defalias 'mm-read-coding-system
88 ((fboundp 'read-coding-system)
89 (if (and (featurep 'xemacs)
90 (<= (string-to-number emacs-version) 21.1))
91 (lambda (prompt &optional default-coding-system)
92 (read-coding-system prompt))
94 (t (lambda (prompt &optional default-coding-system)
95 "Prompt the user for a coding system."
97 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
98 mm-mime-mule-charset-alist)))))))
100 (defvar mm-coding-system-list nil)
101 (defun mm-get-coding-system-list ()
102 "Get the coding system list."
103 (or mm-coding-system-list
104 (setq mm-coding-system-list (mm-coding-system-list))))
106 (defun mm-coding-system-p (sym)
107 "Return non-nil if SYM is a coding system."
108 (or (and (fboundp 'coding-system-p) (coding-system-p sym))
109 (memq sym (mm-get-coding-system-list))))
111 (defvar mm-charset-synonym-alist
113 ;; Perfectly fine? A valid MIME name, anyhow.
114 ,@(unless (mm-coding-system-p 'big5)
116 ;; Not in XEmacs, but it's not a proper MIME charset anyhow.
117 ,@(unless (mm-coding-system-p 'x-ctext)
118 '((x-ctext . ctext)))
119 ;; Apparently not defined in Emacs 20, but is a valid MIME name.
120 ,@(unless (mm-coding-system-p 'gb2312)
121 '((gb2312 . cn-gb-2312)))
122 ;; ISO-8859-15 is very similar to ISO-8859-1.
123 ,@(unless (mm-coding-system-p 'iso-8859-15) ; Emacs 21 defines it.
124 '((iso-8859-15 . iso-8859-1)))
125 ;; Windows-1252 is actually a superset of Latin-1. See also
126 ;; `gnus-article-dumbquotes-map'.
127 ,@(unless (mm-coding-system-p 'windows-1252)
128 (if (mm-coding-system-p 'cp1252)
129 '((windows-1252 . cp1252))
130 '((windows-1252 . iso-8859-1))))
131 ;; Windows-1250 is a variant of Latin-2 heavily used by Microsoft
132 ;; Outlook users in Czech republic. Use this to allow reading of their
133 ;; e-mails. cp1250 should be defined by M-x codepage-setup.
134 ,@(if (and (not (mm-coding-system-p 'windows-1250))
135 (mm-coding-system-p 'cp1250))
136 '((windows-1250 . cp1250)))
138 "A mapping from invalid charset names to the real charset names.")
140 (defvar mm-binary-coding-system
142 ((mm-coding-system-p 'binary) 'binary)
143 ((mm-coding-system-p 'no-conversion) 'no-conversion)
145 "100% binary coding system.")
147 (defvar mm-text-coding-system
148 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
149 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
150 (and (mm-coding-system-p 'raw-text) 'raw-text))
151 mm-binary-coding-system)
152 "Text-safe coding system (For removing ^M).")
154 (defvar mm-text-coding-system-for-write nil
155 "Text coding system for write.")
157 (defvar mm-auto-save-coding-system
159 ((mm-coding-system-p 'utf-8-emacs)
160 (if (memq system-type '(windows-nt ms-dos ms-windows))
161 (if (mm-coding-system-p 'utf-8-emacs-dos)
162 'utf-8-emacs-dos mm-binary-coding-system)
164 ((mm-coding-system-p 'emacs-mule)
165 (if (memq system-type '(windows-nt ms-dos ms-windows))
166 (if (mm-coding-system-p 'emacs-mule-dos)
167 'emacs-mule-dos mm-binary-coding-system)
169 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
170 (t mm-binary-coding-system))
171 "Coding system of auto save file.")
173 (defvar mm-universal-coding-system mm-auto-save-coding-system
174 "The universal coding system.")
176 ;; Fixme: some of the cars here aren't valid MIME charsets. That
177 ;; should only matter with XEmacs, though.
178 (defvar mm-mime-mule-charset-alist
180 (iso-8859-1 latin-iso8859-1)
181 (iso-8859-2 latin-iso8859-2)
182 (iso-8859-3 latin-iso8859-3)
183 (iso-8859-4 latin-iso8859-4)
184 (iso-8859-5 cyrillic-iso8859-5)
185 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
186 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
187 ;; charset is koi8-r, not iso-8859-5.
188 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
189 (iso-8859-6 arabic-iso8859-6)
190 (iso-8859-7 greek-iso8859-7)
191 (iso-8859-8 hebrew-iso8859-8)
192 (iso-8859-9 latin-iso8859-9)
193 (iso-8859-14 latin-iso8859-14)
194 (iso-8859-15 latin-iso8859-15)
195 (viscii vietnamese-viscii-lower)
196 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
197 (euc-kr korean-ksc5601)
198 (gb2312 chinese-gb2312)
199 (big5 chinese-big5-1 chinese-big5-2)
201 (thai-tis620 thai-tis620)
202 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
203 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
204 latin-jisx0201 japanese-jisx0208-1978
205 chinese-gb2312 japanese-jisx0208
206 korean-ksc5601 japanese-jisx0212
208 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
209 latin-jisx0201 japanese-jisx0208-1978
210 chinese-gb2312 japanese-jisx0208
211 korean-ksc5601 japanese-jisx0212
212 chinese-cns11643-1 chinese-cns11643-2)
213 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
214 cyrillic-iso8859-5 greek-iso8859-7
215 latin-jisx0201 japanese-jisx0208-1978
216 chinese-gb2312 japanese-jisx0208
217 korean-ksc5601 japanese-jisx0212
218 chinese-cns11643-1 chinese-cns11643-2
219 chinese-cns11643-3 chinese-cns11643-4
220 chinese-cns11643-5 chinese-cns11643-6
222 ,(if (or (not (fboundp 'charsetp)) ;; non-Mule case
223 (charsetp 'unicode-a)
224 (not (mm-coding-system-p 'mule-utf-8)))
225 '(utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e)
226 ;; If we have utf-8 we're in Mule 5+.
229 (coding-system-get 'mule-utf-8 'safe-charsets)))))
230 "Alist of MIME-charset/MULE-charsets.")
232 ;; Correct by construction, but should be unnecessary:
234 (when (and (not (featurep 'xemacs))
235 (fboundp 'coding-system-list)
236 (fboundp 'sort-coding-systems))
237 (setq mm-mime-mule-charset-alist
242 (when (and (or (coding-system-get cs :mime-charset) ; Emacs 22
243 (coding-system-get cs 'mime-charset))
244 (not (eq t (coding-system-get cs 'safe-charsets))))
245 (list (cons (or (coding-system-get cs :mime-charset)
246 (coding-system-get cs 'mime-charset))
248 (coding-system-get cs 'safe-charsets))))))
249 (sort-coding-systems (coding-system-list 'base-only))))))
251 (defvar mm-coding-system-priorities nil
252 "Preferred coding systems for encoding outgoing mails.
254 More than one suitable coding systems may be found for some texts. By
255 default, a coding system with the highest priority is used to encode
256 outgoing mails (see `sort-coding-systems'). If this variable is set,
257 it overrides the default priority. For example, Japanese users may
258 prefer iso-2022-jp to japanese-shift-jis:
260 \(setq mm-coding-system-priorities
261 '(iso-2022-jp iso-2022-jp-2 japanese-shift-jis utf-8))
264 (defvar mm-use-find-coding-systems-region
265 (fboundp 'find-coding-systems-region)
266 "Use `find-coding-systems-region' to find proper coding systems.")
268 ;;; Internal variables:
272 (defun mm-mule-charset-to-mime-charset (charset)
273 "Return the MIME charset corresponding to the given Mule CHARSET."
274 (if (fboundp 'find-coding-systems-for-charsets)
276 (dolist (cs (find-coding-systems-for-charsets (list charset)))
279 (setq mime (or (coding-system-get cs :mime-charset)
280 (coding-system-get cs 'mime-charset))))))
282 (let ((alist mm-mime-mule-charset-alist)
285 (when (memq charset (cdar alist))
286 (setq out (caar alist)
291 (defun mm-charset-to-coding-system (charset &optional lbt)
292 "Return coding-system corresponding to CHARSET.
293 CHARSET is a symbol naming a MIME charset.
294 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
295 used as the line break code type of the coding system."
296 (when (stringp charset)
297 (setq charset (intern (downcase charset))))
299 (setq charset (intern (format "%s-%s" charset lbt))))
303 ;; Running in a non-MULE environment.
304 ((null (mm-get-coding-system-list))
307 ((eq charset 'us-ascii)
309 ;; Check to see whether we can handle this charset. (This depends
310 ;; on there being some coding system matching each `mime-charset'
311 ;; property defined, as there should be.)
312 ((and (mm-coding-system-p charset)
313 ;;; Doing this would potentially weed out incorrect charsets.
315 ;;; (eq charset (coding-system-get charset 'mime-charset))
318 ;; Translate invalid charsets.
319 ((let ((cs (cdr (assq charset mm-charset-synonym-alist))))
320 (and cs (mm-coding-system-p charset) cs)))
321 ;; Last resort: search the coding system list for entries which
322 ;; have the right mime-charset in case the canonical name isn't
323 ;; defined (though it should be).
325 ;; mm-get-coding-system-list returns a list of cs without lbt.
327 (dolist (c (mm-get-coding-system-list))
329 (eq charset (or (coding-system-get c :mime-charset)
330 (coding-system-get c 'mime-charset))))
334 (defsubst mm-replace-chars-in-string (string from to)
335 (mm-subst-char-in-string from to string))
338 (defvar mm-emacs-mule (and (not (featurep 'xemacs))
339 (boundp 'default-enable-multibyte-characters)
340 default-enable-multibyte-characters
341 (fboundp 'set-buffer-multibyte))
342 "True in Emacs with Mule.")
345 (defun mm-enable-multibyte ()
346 "Set the multibyte flag of the current buffer.
347 Only do this if the default value of `enable-multibyte-characters' is
348 non-nil. This is a no-op in XEmacs."
349 (set-buffer-multibyte t))
350 (defalias 'mm-enable-multibyte 'ignore))
353 (defun mm-disable-multibyte ()
354 "Unset the multibyte flag of in the current buffer.
355 This is a no-op in XEmacs."
356 (set-buffer-multibyte nil))
357 (defalias 'mm-disable-multibyte 'ignore)))
359 (defun mm-preferred-coding-system (charset)
360 ;; A typo in some Emacs versions.
361 (or (get-charset-property charset 'preferred-coding-system)
362 (get-charset-property charset 'prefered-coding-system)))
364 (defun mm-charset-after (&optional pos)
365 "Return charset of a character in current buffer at position POS.
366 If POS is nil, it defauls to the current point.
367 If POS is out of range, the value is nil.
368 If the charset is `composition', return the actual one."
369 (let ((char (char-after pos)) charset)
370 (if (< (mm-char-int char) 128)
371 (setq charset 'ascii)
372 ;; charset-after is fake in some Emacsen.
373 (setq charset (and (fboundp 'char-charset) (char-charset char)))
374 (if (eq charset 'composition) ; Mule 4
375 (let ((p (or pos (point))))
376 (cadr (find-charset-region p (1+ p))))
377 (if (and charset (not (memq charset '(ascii eight-bit-control
378 eight-bit-graphic))))
381 mail-parse-mule-charset ;; cached mule-charset
383 (setq mail-parse-mule-charset
384 (and (boundp 'current-language-environment)
387 (assoc current-language-environment
388 language-info-alist))))))
389 (if (or (not mail-parse-mule-charset)
390 (eq mail-parse-mule-charset 'ascii))
391 (setq mail-parse-mule-charset
392 (or (car (last (assq mail-parse-charset
393 mm-mime-mule-charset-alist)))
394 ;; Fixme: don't fix that!
396 mail-parse-mule-charset)))))))
398 (defun mm-mime-charset (charset)
399 "Return the MIME charset corresponding to the given Mule CHARSET."
400 (if (eq charset 'unknown)
401 (error "The message contains non-printable characters, please use attachment"))
402 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
403 ;; This exists in Emacs 20.
405 (and (mm-preferred-coding-system charset)
406 (or (coding-system-get
407 (mm-preferred-coding-system charset) :mime-charset)
409 (mm-preferred-coding-system charset) 'mime-charset)))
410 (and (eq charset 'ascii)
412 (mm-preferred-coding-system charset)
413 (mm-mule-charset-to-mime-charset charset))
414 ;; This is for XEmacs.
415 (mm-mule-charset-to-mime-charset charset)))
417 (defun mm-delete-duplicates (list)
418 "Simple substitute for CL `delete-duplicates', testing with `equal'."
421 (setq head (car list))
422 (setq list (delete head list))
423 (setq result (cons head result)))
426 ;; It's not clear whether this is supposed to mean the global or local
427 ;; setting. I think it's used inconsistently. -- fx
428 (defsubst mm-multibyte-p ()
429 "Say whether multibyte is enabled."
430 (if (and (not (featurep 'xemacs))
431 (boundp 'enable-multibyte-characters))
432 enable-multibyte-characters
435 (defun mm-sort-coding-systems-predicate (a b)
436 (> (length (memq a mm-coding-system-priorities))
437 (length (memq b mm-coding-system-priorities))))
439 (defun mm-find-mime-charset-region (b e)
440 "Return the MIME charsets needed to encode the region between B and E.
441 nil means ASCII, a single-element list represents an appropriate MIME
442 charset, and a longer list means no appropriate charset."
444 ;; The return possibilities of this function are a mess...
445 (or (and (mm-multibyte-p)
446 mm-use-find-coding-systems-region
447 ;; Find the mime-charset of the most preferred coding
448 ;; system that has one.
449 (let ((systems (find-coding-systems-region b e)))
450 (when mm-coding-system-priorities
452 (sort systems 'mm-sort-coding-systems-predicate)))
453 ;; Fixme: The `mime-charset' (`x-ctext') of `compound-text'
454 ;; is not in the IANA list.
455 (setq systems (delq 'compound-text systems))
456 (unless (equal systems '(undecided))
458 (let* ((head (pop systems))
459 (cs (or (coding-system-get head :mime-charset)
460 (coding-system-get head 'mime-charset))))
463 charsets (list cs))))))
465 ;; Fixme: won't work for unibyte Emacs 22:
467 ;; Otherwise we're not multibyte, XEmacs or a single coding
468 ;; system won't cover it.
470 (mm-delete-duplicates
471 (mapcar 'mm-mime-charset
473 (mm-find-charset-region b e))))))
476 (defmacro mm-with-unibyte-buffer (&rest forms)
477 "Create a temporary buffer, and evaluate FORMS there like `progn'.
478 Use unibyte mode for this."
479 `(let (default-enable-multibyte-characters)
480 (with-temp-buffer ,@forms)))
481 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
482 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
484 (defmacro mm-with-unibyte-current-buffer (&rest forms)
485 "Evaluate FORMS with current buffer temporarily made unibyte.
486 Also bind `default-enable-multibyte-characters' to nil.
487 Equivalent to `progn' in XEmacs"
488 (let ((multibyte (make-symbol "multibyte"))
489 (buffer (make-symbol "buffer")))
491 (let ((,multibyte enable-multibyte-characters)
492 (,buffer (current-buffer)))
494 (let (default-enable-multibyte-characters)
495 (set-buffer-multibyte nil)
498 (set-buffer-multibyte ,multibyte)))
499 (let (default-enable-multibyte-characters)
501 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
502 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
504 (defmacro mm-with-unibyte (&rest forms)
505 "Eval the FORMS with the default value of `enable-multibyte-characters' nil, ."
506 `(let (default-enable-multibyte-characters)
508 (put 'mm-with-unibyte 'lisp-indent-function 0)
509 (put 'mm-with-unibyte 'edebug-form-spec '(body))
511 (defun mm-find-charset-region (b e)
512 "Return a list of Emacs charsets in the region B to E."
514 ((and (mm-multibyte-p)
515 (fboundp 'find-charset-region))
516 ;; Remove composition since the base charsets have been included.
517 ;; Remove eight-bit-*, treat them as ascii.
518 (let ((css (find-charset-region b e)))
519 (mapcar (lambda (cs) (setq css (delq cs css)))
520 '(composition eight-bit-control eight-bit-graphic
524 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
527 (narrow-to-region b e)
528 (goto-char (point-min))
529 (skip-chars-forward "\0-\177")
534 (and (boundp 'current-language-environment)
535 (car (last (assq 'charset
536 (assoc current-language-environment
537 language-info-alist))))))
538 (if (eq charset 'ascii) (setq charset nil))
541 (car (last (assq mail-parse-charset
542 mm-mime-mule-charset-alist)))))
543 (list 'ascii (or charset 'latin-iso8859-1)))))))))
545 (if (fboundp 'shell-quote-argument)
546 (defalias 'mm-quote-arg 'shell-quote-argument)
547 (defun mm-quote-arg (arg)
548 "Return a version of ARG that is safe to evaluate in a shell."
549 (let ((pos 0) new-pos accum)
550 ;; *** bug: we don't handle newline characters properly
551 (while (setq new-pos (string-match "[]*[;!'`\"$\\& \t{} |()<>]" arg pos))
552 (push (substring arg pos new-pos) accum)
554 (push (list (aref arg new-pos)) accum)
555 (setq pos (1+ new-pos)))
558 (apply 'concat (nconc (nreverse accum) (list (substring arg pos))))))))
560 (defun mm-auto-mode-alist ()
561 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
562 (let ((alist auto-mode-alist)
565 (when (listp (cdar alist))
566 (push (car alist) out))
570 (defvar mm-inhibit-file-name-handlers
571 '(jka-compr-handler image-file-handler)
572 "A list of handlers doing (un)compression (etc) thingies.")
574 (defun mm-insert-file-contents (filename &optional visit beg end replace
576 "Like `insert-file-contents', q.v., but only reads in the file.
577 A buffer may be modified in several ways after reading into the buffer due
578 to advanced Emacs features, such as file-name-handlers, format decoding,
579 find-file-hooks, etc.
580 If INHIBIT is non-nil, inhibit `mm-inhibit-file-name-handlers'.
581 This function ensures that none of these modifications will take place."
582 (let ((format-alist nil)
583 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
584 (default-major-mode 'fundamental-mode)
585 (enable-local-variables nil)
586 (after-insert-file-functions nil)
587 (enable-local-eval nil)
588 (find-file-hooks nil)
589 (inhibit-file-name-operation (if inhibit
590 'insert-file-contents
591 inhibit-file-name-operation))
592 (inhibit-file-name-handlers
594 (append mm-inhibit-file-name-handlers
595 inhibit-file-name-handlers)
596 inhibit-file-name-handlers)))
597 (insert-file-contents filename visit beg end replace)))
599 (defun mm-append-to-file (start end filename &optional codesys inhibit)
600 "Append the contents of the region to the end of file FILENAME.
601 When called from a function, expects three arguments,
602 START, END and FILENAME. START and END are buffer positions
603 saying what text to write.
604 Optional fourth argument specifies the coding system to use when
606 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
607 (let ((coding-system-for-write
608 (or codesys mm-text-coding-system-for-write
609 mm-text-coding-system))
610 (inhibit-file-name-operation (if inhibit
612 inhibit-file-name-operation))
613 (inhibit-file-name-handlers
615 (append mm-inhibit-file-name-handlers
616 inhibit-file-name-handlers)
617 inhibit-file-name-handlers)))
618 (append-to-file start end filename)))
620 (defun mm-write-region (start end filename &optional append visit lockname
621 coding-system inhibit)
623 "Like `write-region'.
624 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
625 (let ((coding-system-for-write
626 (or coding-system mm-text-coding-system-for-write
627 mm-text-coding-system))
628 (inhibit-file-name-operation (if inhibit
630 inhibit-file-name-operation))
631 (inhibit-file-name-handlers
633 (append mm-inhibit-file-name-handlers
634 inhibit-file-name-handlers)
635 inhibit-file-name-handlers)))
636 (write-region start end filename append visit lockname)))
638 (defun mm-image-load-path (&optional package)
640 (dolist (path load-path (nreverse result))
641 (if (file-directory-p
642 (setq dir (concat (file-name-directory
643 (directory-file-name path))
644 "etc/" (or package "gnus/"))))
646 (push path result))))
650 ;;; mm-util.el ends here