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 'static))
31 (defvar mm-mime-mule-charset-alist
33 (iso-8859-1 latin-iso8859-1)
34 (iso-8859-2 latin-iso8859-2)
35 (iso-8859-3 latin-iso8859-3)
36 (iso-8859-4 latin-iso8859-4)
37 (iso-8859-5 cyrillic-iso8859-5)
38 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
39 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
40 ;; charset is koi8-r, not iso-8859-5.
41 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
42 (iso-8859-6 arabic-iso8859-6)
43 (iso-8859-7 greek-iso8859-7)
44 (iso-8859-8 hebrew-iso8859-8)
45 (iso-8859-9 latin-iso8859-9)
46 (viscii vietnamese-viscii-lower)
47 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
48 (euc-kr korean-ksc5601)
49 (cn-gb-2312 chinese-gb2312)
50 (cn-big5 chinese-big5-1 chinese-big5-2)
52 (thai-tis620 thai-tis620)
53 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
54 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
55 latin-jisx0201 japanese-jisx0208-1978
56 chinese-gb2312 japanese-jisx0208
57 korean-ksc5601 japanese-jisx0212
59 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
60 latin-jisx0201 japanese-jisx0208-1978
61 chinese-gb2312 japanese-jisx0208
62 korean-ksc5601 japanese-jisx0212
63 chinese-cns11643-1 chinese-cns11643-2)
64 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
65 cyrillic-iso8859-5 greek-iso8859-7
66 latin-jisx0201 japanese-jisx0208-1978
67 chinese-gb2312 japanese-jisx0208
68 korean-ksc5601 japanese-jisx0212
69 chinese-cns11643-1 chinese-cns11643-2
70 chinese-cns11643-3 chinese-cns11643-4
71 chinese-cns11643-5 chinese-cns11643-6
73 (utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
74 "Alist of MIME-charset/MULE-charsets.")
79 (let ((nfunc (intern (format "mm-%s" (car elem)))))
80 (if (fboundp (car elem))
81 (defalias nfunc (car elem))
82 (defalias nfunc (cdr elem)))))
83 '((decode-coding-string . (lambda (s a) s))
84 (encode-coding-string . (lambda (s a) s))
85 (encode-coding-region . ignore)
86 (coding-system-list . ignore)
87 (decode-coding-region . ignore)
89 (device-type . ignore)
90 (coding-system-equal . equal)
91 (annotationp . ignore)
92 (set-buffer-file-coding-system . ignore)
94 . (lambda (charset int)
98 "Prompt the user for a coding system."
100 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
101 mm-mime-mule-charset-alist))))
108 (mapcar (lambda (e) (list (symbol-name (car e))))
109 mm-mime-mule-charset-alist)
113 (defalias 'mm-char-or-char-int-p
115 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
116 ((fboundp 'char-valid-p) 'char-valid-p)
119 (defvar mm-coding-system-list nil)
120 (defun mm-get-coding-system-list ()
121 "Get the coding system list."
122 (or mm-coding-system-list
123 (setq mm-coding-system-list (mm-coding-system-list))))
125 (defvar mm-charset-synonym-alist
127 (gb2312 . cn-gb-2312)
129 "A mapping from invalid charset names to the real charset names.")
131 (defun mm-coding-system-p (sym)
132 "Return non-nil if SYM is a coding system."
133 (or (and (fboundp 'coding-system-p) (coding-system-p sym))
134 (memq sym (mm-get-coding-system-list))))
136 (defvar mm-binary-coding-system
138 ((mm-coding-system-p 'binary) 'binary)
139 ((mm-coding-system-p 'no-conversion) 'no-conversion)
141 "100% binary coding system.")
143 (defvar mm-text-coding-system
144 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
145 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
146 (and (mm-coding-system-p 'raw-text) 'raw-text))
147 mm-binary-coding-system)
148 "Text-safe coding system (For removing ^M).")
150 (defvar mm-text-coding-system-for-write nil
151 "Text coding system for write.")
153 (defvar mm-auto-save-coding-system
155 ((mm-coding-system-p 'emacs-mule)
156 (if (memq system-type '(windows-nt ms-dos ms-windows))
157 (if (mm-coding-system-p 'emacs-mule-dos)
158 'emacs-mule-dos mm-binary-coding-system)
160 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
161 (t mm-binary-coding-system))
162 "Coding system of auto save file.")
164 ;;; Internal variables:
168 (defun mm-mule-charset-to-mime-charset (charset)
169 "Return the MIME charset corresponding to MULE CHARSET."
170 (let ((alist mm-mime-mule-charset-alist)
173 (when (memq charset (cdar alist))
174 (setq out (caar alist)
179 (defun mm-charset-to-coding-system (charset &optional lbt)
180 "Return coding-system corresponding to CHARSET.
181 CHARSET is a symbol naming a MIME charset.
182 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
183 used as the line break code type of the coding system."
184 (when (stringp charset)
185 (setq charset (intern (downcase charset))))
187 (or (cdr (assq charset mm-charset-synonym-alist))
190 (setq charset (intern (format "%s-%s" charset lbt))))
192 ;; Running in a non-MULE environment.
193 ((null (mm-get-coding-system-list))
196 ((eq charset 'us-ascii)
198 ;; Check to see whether we can handle this charset.
199 ((memq charset (mm-get-coding-system-list))
205 (static-if (fboundp 'subst-char-in-string)
206 (defsubst mm-replace-chars-in-string (string from to)
207 (subst-char-in-string from to string))
208 (defun mm-replace-chars-in-string (string from to)
209 "Replace characters in STRING from FROM to TO."
210 (let ((string (substring string 0)) ;Copy string.
211 (len (length string))
213 ;; Replace all occurrences of FROM with TO.
215 (when (= (aref string idx) from)
216 (aset string idx to))
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 (defun mm-preferred-coding-system (charset)
233 ;; A typo in some Emacs versions.
234 (or (get-charset-property charset 'prefered-coding-system)
235 (get-charset-property charset 'preferred-coding-system)))
237 (defun mm-charset-after (&optional pos)
238 "Return charset of a character in current buffer at position POS.
239 If POS is nil, it defauls to the current point.
240 If POS is out of range, the value is nil.
241 If the charset is `composition', return the actual one."
243 ((fboundp 'charset-after)
245 ((fboundp 'char-charset)
246 (char-charset (char-after pos)))
247 ((< (mm-char-int (char-after pos)) 128)
249 (mail-parse-mule-charset ;; cached mule-charset
250 mail-parse-mule-charset)
251 ((boundp 'current-language-environment)
252 (let ((entry (assoc current-language-environment
253 language-info-alist)))
254 (setq mail-parse-mule-charset
255 (or (car (last (assq 'charset entry)))
257 (t ;; figure out the charset
258 (setq mail-parse-mule-charset
259 (or (car (last (assq mail-parse-charset
260 mm-mime-mule-charset-alist)))
261 'latin-iso8859-1))))))
262 (if (eq charset 'composition)
263 (let ((p (or pos (point))))
264 (cadr (find-charset-region p (1+ p))))
267 (defun mm-mime-charset (charset)
268 "Return the MIME charset corresponding to the MULE CHARSET."
269 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
270 ;; This exists in Emacs 20.
272 (and (mm-preferred-coding-system charset)
274 (mm-preferred-coding-system charset) 'mime-charset))
275 (and (eq charset 'ascii)
277 (mm-preferred-coding-system charset)
278 (mm-mule-charset-to-mime-charset charset))
279 ;; This is for XEmacs.
280 (mm-mule-charset-to-mime-charset charset)))
282 (defun mm-delete-duplicates (list)
283 "Simple substitute for CL `delete-duplicates', testing with `equal'."
286 (setq head (car list))
287 (setq list (delete head list))
288 (setq result (cons head result)))
291 (defun mm-find-mime-charset-region (b e)
292 "Return the MIME charsets needed to encode the region between B and E."
293 (let ((charsets (mapcar 'mm-mime-charset
295 (mm-find-charset-region b e)))))
296 (when (memq 'iso-2022-jp-2 charsets)
297 (setq charsets (delq 'iso-2022-jp charsets)))
298 (setq charsets (mm-delete-duplicates charsets))
299 (if (and (> (length charsets) 1)
300 (fboundp 'find-coding-systems-region)
301 (memq 'utf-8 (find-coding-systems-region b e)))
305 (defsubst mm-multibyte-p ()
306 "Say whether multibyte is enabled."
307 (or (featurep 'xemacs)
308 (and (boundp 'enable-multibyte-characters)
309 enable-multibyte-characters)))
311 (defmacro mm-with-unibyte-buffer (&rest forms)
312 "Create a temporary buffer, and evaluate FORMS there like `progn'.
313 See also `with-temp-file' and `with-output-to-string'."
314 (let ((temp-buffer (make-symbol "temp-buffer"))
315 (multibyte (make-symbol "multibyte")))
316 `(if (or (string-match "XEmacs\\|Lucid" emacs-version)
317 (not (boundp 'enable-multibyte-characters)))
318 (with-temp-buffer ,@forms)
319 (let ((,multibyte (default-value 'enable-multibyte-characters))
323 (setq-default enable-multibyte-characters nil)
325 (get-buffer-create (generate-new-buffer-name " *temp*")))
327 (with-current-buffer ,temp-buffer
328 (let ((buffer-file-coding-system mm-binary-coding-system)
329 (coding-system-for-read mm-binary-coding-system)
330 (coding-system-for-write mm-binary-coding-system))
332 (and (buffer-name ,temp-buffer)
333 (kill-buffer ,temp-buffer))))
334 (setq-default enable-multibyte-characters ,multibyte))))))
335 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
336 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
338 (defmacro mm-with-unibyte-current-buffer (&rest forms)
339 "Evaluate FORMS there like `progn' in current buffer."
340 (let ((multibyte (make-symbol "multibyte")))
341 `(if (or (featurep 'xemacs)
342 (not (fboundp 'set-buffer-multibyte)))
345 (let ((,multibyte (default-value 'enable-multibyte-characters)))
347 (let ((buffer-file-coding-system mm-binary-coding-system)
348 (coding-system-for-read mm-binary-coding-system)
349 (coding-system-for-write mm-binary-coding-system))
350 (set-buffer-multibyte nil)
351 (setq-default enable-multibyte-characters nil)
353 (setq-default enable-multibyte-characters ,multibyte)
354 (set-buffer-multibyte ,multibyte))))))
355 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
356 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
358 (defmacro mm-with-unibyte (&rest forms)
359 "Set default `enable-multibyte-characters' to `nil', eval the FORMS."
360 (let ((multibyte (make-symbol "multibyte")))
361 `(if (or (featurep 'xemacs)
362 (not (boundp 'enable-multibyte-characters)))
364 (let ((,multibyte (default-value 'enable-multibyte-characters)))
367 (setq-default enable-multibyte-characters nil)
369 (setq-default enable-multibyte-characters ,multibyte))))))
370 (put 'mm-with-unibyte 'lisp-indent-function 0)
371 (put 'mm-with-unibyte 'edebug-form-spec '(body))
373 (defun mm-find-charset-region (b e)
374 "Return a list of charsets in the region."
376 ((and (mm-multibyte-p)
377 (fboundp 'find-charset-region))
378 ;; Remove composition since the base charsets have been included.
379 (delq 'composition (find-charset-region b e)))
380 ((not (boundp 'current-language-environment))
383 (narrow-to-region b e)
384 (goto-char (point-min))
385 (skip-chars-forward "\0-\177")
388 (delq nil (list 'ascii
389 (or (car (last (assq mail-parse-charset
390 mm-mime-mule-charset-alist)))
391 'latin-iso8859-1)))))))
393 ;; We are in a unibyte buffer, so we futz around a bit.
396 (narrow-to-region b e)
397 (goto-char (point-min))
398 (let ((entry (assoc current-language-environment
399 language-info-alist)))
400 (skip-chars-forward "\0-\177")
403 (delq nil (list 'ascii
404 (or (car (last (assq 'charset entry)))
405 'latin-iso8859-1))))))))))
407 (static-if (fboundp 'shell-quote-argument)
408 (defalias 'mm-quote-arg 'shell-quote-argument)
409 (defun mm-quote-arg (arg)
410 "Return a version of ARG that is safe to evaluate in a shell."
411 (let ((pos 0) new-pos accum)
412 ;; *** bug: we don't handle newline characters properly
413 (while (setq new-pos (string-match "[]*[;!'`\"$\\& \t{} |()<>]" arg pos))
414 (push (substring arg pos new-pos) accum)
416 (push (list (aref arg new-pos)) accum)
417 (setq pos (1+ new-pos)))
420 (apply 'concat (nconc (nreverse accum) (list (substring arg pos))))))))
422 (defun mm-auto-mode-alist ()
423 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
424 (let ((alist auto-mode-alist)
427 (when (listp (cdar alist))
428 (push (car alist) out))
432 (defvar mm-inhibit-file-name-handlers
434 "A list of handlers doing (un)compression (etc) thingies.")
436 (defun mm-insert-file-contents (filename &optional visit beg end replace
438 "Like `insert-file-contents', q.v., but only reads in the file.
439 A buffer may be modified in several ways after reading into the buffer due
440 to advanced Emacs features, such as file-name-handlers, format decoding,
441 find-file-hooks, etc.
442 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers.
443 This function ensures that none of these modifications will take place."
444 (let ((format-alist nil)
445 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
446 (default-major-mode 'fundamental-mode)
447 (enable-local-variables nil)
448 (after-insert-file-functions nil)
449 (enable-local-eval nil)
450 (find-file-hooks nil)
451 (inhibit-file-name-operation (if inhibit
452 'insert-file-contents
453 inhibit-file-name-operation))
454 (inhibit-file-name-handlers
456 (append mm-inhibit-file-name-handlers
457 inhibit-file-name-handlers)
458 inhibit-file-name-handlers)))
459 (insert-file-contents filename visit beg end replace)))
461 (defun mm-append-to-file (start end filename &optional codesys inhibit)
462 "Append the contents of the region to the end of file FILENAME.
463 When called from a function, expects three arguments,
464 START, END and FILENAME. START and END are buffer positions
465 saying what text to write.
466 Optional fourth argument specifies the coding system to use when
468 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
469 (let ((coding-system-for-write
470 (or codesys mm-text-coding-system-for-write
471 mm-text-coding-system))
472 (inhibit-file-name-operation (if inhibit
474 inhibit-file-name-operation))
475 (inhibit-file-name-handlers
477 (append mm-inhibit-file-name-handlers
478 inhibit-file-name-handlers)
479 inhibit-file-name-handlers)))
480 (append-to-file start end filename)))
482 (defun mm-write-region (start end filename &optional append visit lockname
483 coding-system inhibit)
485 "Like `write-region'.
486 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
487 (let ((coding-system-for-write
488 (or coding-system mm-text-coding-system-for-write
489 mm-text-coding-system))
490 (inhibit-file-name-operation (if inhibit
492 inhibit-file-name-operation))
493 (inhibit-file-name-handlers
495 (append mm-inhibit-file-name-handlers
496 inhibit-file-name-handlers)
497 inhibit-file-name-handlers)))
498 (write-region start end filename append visit lockname)))
502 ;;; mm-util.el ends here