1 ;;; mm-util.el --- Utility functions for MIME things
2 ;; Copyright (C) 1998, 1999, 2000 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.
27 (eval-when-compile (require 'cl))
30 (defvar mm-mime-mule-charset-alist
32 (iso-8859-1 latin-iso8859-1)
33 (iso-8859-2 latin-iso8859-2)
34 (iso-8859-3 latin-iso8859-3)
35 (iso-8859-4 latin-iso8859-4)
36 (iso-8859-5 cyrillic-iso8859-5)
37 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
38 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
39 ;; charset is koi8-r, not iso-8859-5.
40 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
41 (iso-8859-6 arabic-iso8859-6)
42 (iso-8859-7 greek-iso8859-7)
43 (iso-8859-8 hebrew-iso8859-8)
44 (iso-8859-9 latin-iso8859-9)
45 (viscii vietnamese-viscii-lower)
46 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
47 (euc-kr korean-ksc5601)
48 (cn-gb-2312 chinese-gb2312)
49 (cn-big5 chinese-big5-1 chinese-big5-2)
51 (thai-tis620 thai-tis620)
52 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
53 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
54 latin-jisx0201 japanese-jisx0208-1978
55 chinese-gb2312 japanese-jisx0208
56 korean-ksc5601 japanese-jisx0212
58 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
59 latin-jisx0201 japanese-jisx0208-1978
60 chinese-gb2312 japanese-jisx0208
61 korean-ksc5601 japanese-jisx0212
62 chinese-cns11643-1 chinese-cns11643-2)
63 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
64 cyrillic-iso8859-5 greek-iso8859-7
65 latin-jisx0201 japanese-jisx0208-1978
66 chinese-gb2312 japanese-jisx0208
67 korean-ksc5601 japanese-jisx0212
68 chinese-cns11643-1 chinese-cns11643-2
69 chinese-cns11643-3 chinese-cns11643-4
70 chinese-cns11643-5 chinese-cns11643-6
72 (utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
73 "Alist of MIME-charset/MULE-charsets.")
78 (let ((nfunc (intern (format "mm-%s" (car elem)))))
79 (if (fboundp (car elem))
80 (defalias nfunc (car elem))
81 (defalias nfunc (cdr elem)))))
82 '((decode-coding-string . (lambda (s a) s))
83 (encode-coding-string . (lambda (s a) s))
84 (encode-coding-region . ignore)
85 (coding-system-list . ignore)
86 (decode-coding-region . ignore)
88 (device-type . ignore)
89 (coding-system-equal . equal)
90 (annotationp . ignore)
91 (set-buffer-file-coding-system . ignore)
93 . (lambda (charset int)
97 "Prompt the user for a coding system."
99 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
100 mm-mime-mule-charset-alist))))
107 (mapcar (lambda (e) (list (symbol-name (car e))))
108 mm-mime-mule-charset-alist)
110 (subst-char-in-string
111 . (lambda (from to string) ;; stolen (and renamed) from nnheader.el
112 "Replace characters in STRING from FROM to TO."
113 (let ((string (substring string 0)) ;Copy string.
114 (len (length string))
116 ;; Replace all occurrences of FROM with TO.
118 (when (= (aref string idx) from)
119 (aset string idx to))
125 (defalias 'mm-char-or-char-int-p
127 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
128 ((fboundp 'char-valid-p) 'char-valid-p)
131 (defvar mm-coding-system-list nil)
132 (defun mm-get-coding-system-list ()
133 "Get the coding system list."
134 (or mm-coding-system-list
135 (setq mm-coding-system-list (mm-coding-system-list))))
137 (defvar mm-charset-synonym-alist
139 (gb2312 . cn-gb-2312)
141 "A mapping from invalid charset names to the real charset names.")
143 (defun mm-coding-system-p (sym)
144 "Return non-nil if SYM is a coding system."
145 (or (and (fboundp 'coding-system-p) (coding-system-p sym))
146 (memq sym (mm-get-coding-system-list))))
148 (defvar mm-binary-coding-system
150 ((mm-coding-system-p 'binary) 'binary)
151 ((mm-coding-system-p 'no-conversion) 'no-conversion)
153 "100% binary coding system.")
155 (defvar mm-text-coding-system
156 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
157 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
158 (and (mm-coding-system-p 'raw-text) 'raw-text))
159 mm-binary-coding-system)
160 "Text-safe coding system (For removing ^M).")
162 (defvar mm-text-coding-system-for-write nil
163 "Text coding system for write.")
165 (defvar mm-auto-save-coding-system
167 ((mm-coding-system-p 'emacs-mule)
168 (if (memq system-type '(windows-nt ms-dos ms-windows))
169 (if (mm-coding-system-p 'emacs-mule-dos)
170 'emacs-mule-dos mm-binary-coding-system)
172 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
173 (t mm-binary-coding-system))
174 "Coding system of auto save file.")
176 ;;; Internal variables:
180 (defun mm-mule-charset-to-mime-charset (charset)
181 "Return the MIME charset corresponding to MULE CHARSET."
182 (let ((alist mm-mime-mule-charset-alist)
185 (when (memq charset (cdar alist))
186 (setq out (caar alist)
191 (defun mm-charset-to-coding-system (charset &optional lbt)
192 "Return coding-system corresponding to CHARSET.
193 CHARSET is a symbol naming a MIME charset.
194 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
195 used as the line break code type of the coding system."
196 (when (stringp charset)
197 (setq charset (intern (downcase charset))))
199 (or (cdr (assq charset mm-charset-synonym-alist))
202 (setq charset (intern (format "%s-%s" charset lbt))))
204 ;; Running in a non-MULE environment.
205 ((null (mm-get-coding-system-list))
208 ((eq charset 'us-ascii)
210 ;; Check to see whether we can handle this charset.
211 ((memq charset (mm-get-coding-system-list))
217 (defsubst mm-replace-chars-in-string (string from to)
218 (mm-subst-char-in-string from to string))
220 (defsubst mm-enable-multibyte ()
221 "Enable multibyte in the current buffer."
222 (when (and (fboundp 'set-buffer-multibyte)
223 (boundp 'enable-multibyte-characters)
224 (default-value 'enable-multibyte-characters))
225 (set-buffer-multibyte t)))
227 (defsubst mm-disable-multibyte ()
228 "Disable multibyte in the current buffer."
229 (when (fboundp 'set-buffer-multibyte)
230 (set-buffer-multibyte nil)))
232 (defsubst mm-enable-multibyte-mule4 ()
233 "Enable multibyte in the current buffer.
234 Only used in Emacs Mule 4."
235 (when (and (fboundp 'set-buffer-multibyte)
236 (boundp 'enable-multibyte-characters)
237 (default-value 'enable-multibyte-characters)
238 (not (charsetp 'eight-bit-control)))
239 (set-buffer-multibyte t)))
241 (defsubst mm-disable-multibyte-mule4 ()
242 "Disable multibyte in the current buffer.
243 Only used in Emacs Mule 4."
244 (when (and (fboundp 'set-buffer-multibyte)
245 (not (charsetp 'eight-bit-control)))
246 (set-buffer-multibyte nil)))
248 (defun mm-preferred-coding-system (charset)
249 ;; A typo in some Emacs versions.
250 (or (get-charset-property charset 'prefered-coding-system)
251 (get-charset-property charset 'preferred-coding-system)))
253 (defun mm-charset-after (&optional pos)
254 "Return charset of a character in current buffer at position POS.
255 If POS is nil, it defauls to the current point.
256 If POS is out of range, the value is nil.
257 If the charset is `composition', return the actual one."
258 (let ((char (char-after pos)) charset)
259 (if (< (mm-char-int char) 128)
260 (setq charset 'ascii)
261 ;; charset-after is fake in some Emacsen.
262 (setq charset (and (fboundp 'char-charset) (char-charset char)))
263 (if (eq charset 'composition)
264 (let ((p (or pos (point))))
265 (cadr (find-charset-region p (1+ p))))
266 (if (and charset (not (memq charset '(ascii eight-bit-control
267 eight-bit-graphic))))
270 mail-parse-mule-charset ;; cached mule-charset
272 (setq mail-parse-mule-charset
273 (and (boundp 'current-language-environment)
276 (assoc current-language-environment
277 language-info-alist))))))
278 (if (or (not mail-parse-mule-charset)
279 (eq mail-parse-mule-charset 'ascii))
280 (setq mail-parse-mule-charset
281 (or (car (last (assq mail-parse-charset
282 mm-mime-mule-charset-alist)))
284 mail-parse-mule-charset)))))))
286 (defun mm-mime-charset (charset)
287 "Return the MIME charset corresponding to the MULE CHARSET."
288 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
289 ;; This exists in Emacs 20.
291 (and (mm-preferred-coding-system charset)
293 (mm-preferred-coding-system charset) 'mime-charset))
294 (and (eq charset 'ascii)
296 (mm-preferred-coding-system charset)
297 (mm-mule-charset-to-mime-charset charset))
298 ;; This is for XEmacs.
299 (mm-mule-charset-to-mime-charset charset)))
301 (defun mm-delete-duplicates (list)
302 "Simple substitute for CL `delete-duplicates', testing with `equal'."
305 (setq head (car list))
306 (setq list (delete head list))
307 (setq result (cons head result)))
310 (defun mm-find-mime-charset-region (b e)
311 "Return the MIME charsets needed to encode the region between B and E."
312 (let ((charsets (mapcar 'mm-mime-charset
314 (mm-find-charset-region b e)))))
315 (when (memq 'iso-2022-jp-2 charsets)
316 (setq charsets (delq 'iso-2022-jp charsets)))
317 (setq charsets (mm-delete-duplicates charsets))
318 (if (and (> (length charsets) 1)
319 (fboundp 'find-coding-systems-region)
320 (memq 'utf-8 (find-coding-systems-region b e)))
324 (defsubst mm-multibyte-p ()
325 "Say whether multibyte is enabled."
326 (if (and (not (featurep 'xemacs))
327 (boundp 'enable-multibyte-characters))
328 enable-multibyte-characters
331 (defmacro mm-with-unibyte-buffer (&rest forms)
332 "Create a temporary buffer, and evaluate FORMS there like `progn'.
333 See also `with-temp-file' and `with-output-to-string'."
334 (let ((temp-buffer (make-symbol "temp-buffer"))
335 (multibyte (make-symbol "multibyte")))
336 `(if (or (featurep 'xemacs)
337 (not (boundp 'enable-multibyte-characters)))
338 (with-temp-buffer ,@forms)
339 (let ((,multibyte (default-value 'enable-multibyte-characters))
343 (setq-default enable-multibyte-characters nil)
345 (get-buffer-create (generate-new-buffer-name " *temp*")))
347 (with-current-buffer ,temp-buffer
348 (let ((buffer-file-coding-system mm-binary-coding-system)
349 (coding-system-for-read mm-binary-coding-system)
350 (coding-system-for-write mm-binary-coding-system))
352 (and (buffer-name ,temp-buffer)
353 (kill-buffer ,temp-buffer))))
354 (setq-default enable-multibyte-characters ,multibyte))))))
355 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
356 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
358 (defmacro mm-with-unibyte-current-buffer (&rest forms)
359 "Evaluate FORMS there like `progn' in current buffer."
360 (let ((multibyte (make-symbol "multibyte")))
361 `(if (or (featurep 'xemacs)
362 (not (fboundp 'set-buffer-multibyte)))
365 (let ((,multibyte (default-value 'enable-multibyte-characters)))
367 (let ((buffer-file-coding-system mm-binary-coding-system)
368 (coding-system-for-read mm-binary-coding-system)
369 (coding-system-for-write mm-binary-coding-system))
370 (set-buffer-multibyte nil)
371 (setq-default enable-multibyte-characters nil)
373 (setq-default enable-multibyte-characters ,multibyte)
374 (set-buffer-multibyte ,multibyte))))))
375 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
376 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
378 (defmacro mm-with-unibyte-current-buffer-mule4 (&rest forms)
379 "Evaluate FORMS there like `progn' in current buffer.
381 (let ((multibyte (make-symbol "multibyte")))
382 `(if (or (featurep 'xemacs)
383 (not (fboundp 'set-buffer-multibyte))
384 (charsetp 'eight-bit-control)) ;; For Emacs Mule 4 only.
387 (let ((,multibyte (default-value 'enable-multibyte-characters)))
389 (let ((buffer-file-coding-system mm-binary-coding-system)
390 (coding-system-for-read mm-binary-coding-system)
391 (coding-system-for-write mm-binary-coding-system))
392 (set-buffer-multibyte nil)
393 (setq-default enable-multibyte-characters nil)
395 (setq-default enable-multibyte-characters ,multibyte)
396 (set-buffer-multibyte ,multibyte))))))
397 (put 'mm-with-unibyte-current-buffer-mule4 'lisp-indent-function 0)
398 (put 'mm-with-unibyte-current-buffer-mule4 'edebug-form-spec '(body))
400 (defmacro mm-with-unibyte (&rest forms)
401 "Set default `enable-multibyte-characters' to `nil', eval the FORMS."
402 (let ((multibyte (make-symbol "multibyte")))
403 `(if (or (featurep 'xemacs)
404 (not (boundp 'enable-multibyte-characters)))
406 (let ((,multibyte (default-value 'enable-multibyte-characters)))
409 (setq-default enable-multibyte-characters nil)
411 (setq-default enable-multibyte-characters ,multibyte))))))
412 (put 'mm-with-unibyte 'lisp-indent-function 0)
413 (put 'mm-with-unibyte 'edebug-form-spec '(body))
415 (defun mm-find-charset-region (b e)
416 "Return a list of charsets in the region."
418 ((and (mm-multibyte-p)
419 (fboundp 'find-charset-region))
420 ;; Remove composition since the base charsets have been included.
421 (delq 'composition (find-charset-region b e)))
423 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
426 (narrow-to-region b e)
427 (goto-char (point-min))
428 (skip-chars-forward "\0-\177")
433 (and (boundp 'current-language-environment)
434 (car (last (assq 'charset
435 (assoc current-language-environment
436 language-info-alist))))))
437 (if (eq charset 'ascii) (setq charset nil))
440 (car (last (assq mail-parse-charset
441 mm-mime-mule-charset-alist)))))
442 (list 'ascii (or charset 'latin-iso8859-1)))))))))
444 (if (fboundp 'shell-quote-argument)
445 (defalias 'mm-quote-arg 'shell-quote-argument)
446 (defun mm-quote-arg (arg)
447 "Return a version of ARG that is safe to evaluate in a shell."
448 (let ((pos 0) new-pos accum)
449 ;; *** bug: we don't handle newline characters properly
450 (while (setq new-pos (string-match "[]*[;!'`\"$\\& \t{} |()<>]" arg pos))
451 (push (substring arg pos new-pos) accum)
453 (push (list (aref arg new-pos)) accum)
454 (setq pos (1+ new-pos)))
457 (apply 'concat (nconc (nreverse accum) (list (substring arg pos))))))))
459 (defun mm-auto-mode-alist ()
460 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
461 (let ((alist auto-mode-alist)
464 (when (listp (cdar alist))
465 (push (car alist) out))
469 (defvar mm-inhibit-file-name-handlers
471 "A list of handlers doing (un)compression (etc) thingies.")
473 (defun mm-insert-file-contents (filename &optional visit beg end replace
475 "Like `insert-file-contents', q.v., but only reads in the file.
476 A buffer may be modified in several ways after reading into the buffer due
477 to advanced Emacs features, such as file-name-handlers, format decoding,
478 find-file-hooks, etc.
479 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers.
480 This function ensures that none of these modifications will take place."
481 (let ((format-alist nil)
482 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
483 (default-major-mode 'fundamental-mode)
484 (enable-local-variables nil)
485 (after-insert-file-functions nil)
486 (enable-local-eval nil)
487 (find-file-hooks nil)
488 (inhibit-file-name-operation (if inhibit
489 'insert-file-contents
490 inhibit-file-name-operation))
491 (inhibit-file-name-handlers
493 (append mm-inhibit-file-name-handlers
494 inhibit-file-name-handlers)
495 inhibit-file-name-handlers)))
496 (insert-file-contents filename visit beg end replace)))
498 (defun mm-append-to-file (start end filename &optional codesys inhibit)
499 "Append the contents of the region to the end of file FILENAME.
500 When called from a function, expects three arguments,
501 START, END and FILENAME. START and END are buffer positions
502 saying what text to write.
503 Optional fourth argument specifies the coding system to use when
505 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
506 (let ((coding-system-for-write
507 (or codesys mm-text-coding-system-for-write
508 mm-text-coding-system))
509 (inhibit-file-name-operation (if inhibit
511 inhibit-file-name-operation))
512 (inhibit-file-name-handlers
514 (append mm-inhibit-file-name-handlers
515 inhibit-file-name-handlers)
516 inhibit-file-name-handlers)))
517 (append-to-file start end filename)))
519 (defun mm-write-region (start end filename &optional append visit lockname
520 coding-system inhibit)
522 "Like `write-region'.
523 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
524 (let ((coding-system-for-write
525 (or coding-system mm-text-coding-system-for-write
526 mm-text-coding-system))
527 (inhibit-file-name-operation (if inhibit
529 inhibit-file-name-operation))
530 (inhibit-file-name-handlers
532 (append mm-inhibit-file-name-handlers
533 inhibit-file-name-handlers)
534 inhibit-file-name-handlers)))
535 (write-region start end filename append visit lockname)))
539 ;;; mm-util.el ends here