1 ;;; mm-util.el --- Utility functions for Mule and low level things
2 ;; Copyright (C) 1998, 1999, 2000, 2001 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))
29 (eval-when-compile (require 'cl))
32 (defvar mm-mime-mule-charset-alist
34 (iso-8859-1 latin-iso8859-1)
35 (iso-8859-2 latin-iso8859-2)
36 (iso-8859-3 latin-iso8859-3)
37 (iso-8859-4 latin-iso8859-4)
38 (iso-8859-5 cyrillic-iso8859-5)
39 ;; Non-mule (X)Emacs uses the last mule-charset for 8bit characters.
40 ;; The fake mule-charset, gnus-koi8-r, tells Gnus that the default
41 ;; charset is koi8-r, not iso-8859-5.
42 (koi8-r cyrillic-iso8859-5 gnus-koi8-r)
43 (iso-8859-6 arabic-iso8859-6)
44 (iso-8859-7 greek-iso8859-7)
45 (iso-8859-8 hebrew-iso8859-8)
46 (iso-8859-9 latin-iso8859-9)
47 (iso-8859-14 latin-iso8859-14)
48 (iso-8859-15 latin-iso8859-15)
49 (viscii vietnamese-viscii-lower)
50 (iso-2022-jp latin-jisx0201 japanese-jisx0208 japanese-jisx0208-1978)
51 (euc-kr korean-ksc5601)
52 (cn-gb-2312 chinese-gb2312)
53 (cn-big5 chinese-big5-1 chinese-big5-2)
55 (thai-tis620 thai-tis620)
56 (iso-2022-7bit ethiopic arabic-1-column arabic-2-column)
57 (iso-2022-jp-2 latin-iso8859-1 greek-iso8859-7
58 latin-jisx0201 japanese-jisx0208-1978
59 chinese-gb2312 japanese-jisx0208
60 korean-ksc5601 japanese-jisx0212
62 (iso-2022-int-1 latin-iso8859-1 greek-iso8859-7
63 latin-jisx0201 japanese-jisx0208-1978
64 chinese-gb2312 japanese-jisx0208
65 korean-ksc5601 japanese-jisx0212
66 chinese-cns11643-1 chinese-cns11643-2)
67 (iso-2022-int-1 latin-iso8859-1 latin-iso8859-2
68 cyrillic-iso8859-5 greek-iso8859-7
69 latin-jisx0201 japanese-jisx0208-1978
70 chinese-gb2312 japanese-jisx0208
71 korean-ksc5601 japanese-jisx0212
72 chinese-cns11643-1 chinese-cns11643-2
73 chinese-cns11643-3 chinese-cns11643-4
74 chinese-cns11643-5 chinese-cns11643-6
76 (utf-8 unicode-a unicode-b unicode-c unicode-d unicode-e))
77 "Alist of MIME-charset/MULE-charsets.")
82 (let ((nfunc (intern (format "mm-%s" (car elem)))))
83 (if (fboundp (car elem))
84 (defalias nfunc (car elem))
85 (defalias nfunc (cdr elem)))))
86 '((decode-coding-string . (lambda (s a) s))
87 (encode-coding-string . (lambda (s a) s))
88 (encode-coding-region . ignore)
89 (coding-system-list . ignore)
90 (decode-coding-region . ignore)
92 (device-type . ignore)
93 (coding-system-equal . equal)
94 (annotationp . ignore)
95 (set-buffer-file-coding-system . ignore)
97 . (lambda (charset int)
101 "Prompt the user for a coding system."
103 prompt (mapcar (lambda (s) (list (symbol-name (car s))))
104 mm-mime-mule-charset-alist))))
111 (mapcar (lambda (e) (list (symbol-name (car e))))
112 mm-mime-mule-charset-alist)
114 (subst-char-in-string
115 . (lambda (from to string) ;; stolen (and renamed) from nnheader.el
116 "Replace characters in STRING from FROM to TO."
117 (let ((string (substring string 0)) ;Copy string.
118 (len (length string))
120 ;; Replace all occurrences of FROM with TO.
122 (when (= (aref string idx) from)
123 (aset string idx to))
126 (string-as-unibyte . identity)
130 (defalias 'mm-char-or-char-int-p
132 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
133 ((fboundp 'char-valid-p) 'char-valid-p)
136 (defvar mm-coding-system-list nil)
137 (defun mm-get-coding-system-list ()
138 "Get the coding system list."
139 (or mm-coding-system-list
140 (setq mm-coding-system-list (mm-coding-system-list))))
142 (defun mm-coding-system-p (sym)
143 "Return non-nil if SYM is a coding system."
144 (or (and (fboundp 'coding-system-p) (coding-system-p sym))
145 (memq sym (mm-get-coding-system-list))))
147 (defvar mm-charset-synonym-alist
149 (gb2312 . cn-gb-2312)
151 ;; Windows-1252 is actually a superset of Latin-1. See also
152 ;; `gnus-article-dumbquotes-map'.
153 ,(unless (mm-coding-system-p 'windows-1252) ; should be defined eventually
154 '(windows-1252 . iso-8859-1))
156 "A mapping from invalid charset names to the real charset names.")
158 (defvar mm-binary-coding-system
160 ((mm-coding-system-p 'binary) 'binary)
161 ((mm-coding-system-p 'no-conversion) 'no-conversion)
163 "100% binary coding system.")
165 (defvar mm-text-coding-system
166 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
167 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
168 (and (mm-coding-system-p 'raw-text) 'raw-text))
169 mm-binary-coding-system)
170 "Text-safe coding system (For removing ^M).")
172 (defvar mm-text-coding-system-for-write nil
173 "Text coding system for write.")
175 (defvar mm-auto-save-coding-system
177 ((mm-coding-system-p 'emacs-mule)
178 (if (memq system-type '(windows-nt ms-dos ms-windows))
179 (if (mm-coding-system-p 'emacs-mule-dos)
180 'emacs-mule-dos mm-binary-coding-system)
182 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
183 (t mm-binary-coding-system))
184 "Coding system of auto save file.")
186 ;;; Internal variables:
190 (defun mm-mule-charset-to-mime-charset (charset)
191 "Return the MIME charset corresponding to the given Mule CHARSET."
192 (let ((alist mm-mime-mule-charset-alist)
195 (when (memq charset (cdar alist))
196 (setq out (caar alist)
201 (defun mm-charset-to-coding-system (charset &optional lbt)
202 "Return coding-system corresponding to CHARSET.
203 CHARSET is a symbol naming a MIME charset.
204 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
205 used as the line break code type of the coding system."
206 (when (stringp charset)
207 (setq charset (intern (downcase charset))))
209 (or (cdr (assq charset mm-charset-synonym-alist))
212 (setq charset (intern (format "%s-%s" charset lbt))))
214 ;; Running in a non-MULE environment.
215 ((null (mm-get-coding-system-list))
218 ((eq charset 'us-ascii)
220 ;; Check to see whether we can handle this charset. (This depends
221 ;; on there being some coding system matching each `mime-charset'
222 ;; coding sysytem property defined, as there should be.)
223 ((memq charset (mm-get-coding-system-list))
229 (defsubst mm-replace-chars-in-string (string from to)
230 (mm-subst-char-in-string from to string))
232 (defsubst mm-enable-multibyte ()
233 "Set the multibyte flag of the current buffer.
234 Only do this if the default value of `enable-multibyte-characters' is
235 non-nil. This is a no-op in XEmacs."
236 (when (and (not (featurep 'xemacs))
237 (boundp 'default-enable-multibyte-characters)
238 default-enable-multibyte-characters
239 (fboundp 'set-buffer-multibyte))
240 (set-buffer-multibyte t)))
242 (defsubst mm-disable-multibyte ()
243 "Unset the multibyte flag of in the current buffer.
244 This is a no-op in XEmacs."
245 (when (and (not (featurep 'xemacs))
246 (fboundp 'set-buffer-multibyte))
247 (set-buffer-multibyte nil)))
249 (defsubst mm-enable-multibyte-mule4 ()
250 "Enable multibyte in the current buffer.
251 Only used in Emacs Mule 4."
252 (when (and (not (featurep 'xemacs))
253 (boundp 'default-enable-multibyte-characters)
254 default-enable-multibyte-characters
255 (fboundp 'set-buffer-multibyte)
257 (not (charsetp 'eight-bit-control)))
258 (set-buffer-multibyte t)))
260 (defsubst mm-disable-multibyte-mule4 ()
261 "Disable multibyte in the current buffer.
262 Only used in Emacs Mule 4."
263 (when (and (not (featurep 'xemacs))
264 (fboundp 'set-buffer-multibyte)
266 (not (charsetp 'eight-bit-control)))
267 (set-buffer-multibyte nil)))
269 (defun mm-preferred-coding-system (charset)
270 ;; A typo in some Emacs versions.
271 (or (get-charset-property charset 'prefered-coding-system)
272 (get-charset-property charset 'preferred-coding-system)))
274 (defun mm-charset-after (&optional pos)
275 "Return charset of a character in current buffer at position POS.
276 If POS is nil, it defauls to the current point.
277 If POS is out of range, the value is nil.
278 If the charset is `composition', return the actual one."
279 (let ((char (char-after pos)) charset)
280 (if (< (mm-char-int char) 128)
281 (setq charset 'ascii)
282 ;; charset-after is fake in some Emacsen.
283 (setq charset (and (fboundp 'char-charset) (char-charset char)))
284 (if (eq charset 'composition)
285 (let ((p (or pos (point))))
286 (cadr (find-charset-region p (1+ p))))
287 (if (and charset (not (memq charset '(ascii eight-bit-control
288 eight-bit-graphic))))
291 mail-parse-mule-charset ;; cached mule-charset
293 (setq mail-parse-mule-charset
294 (and (boundp 'current-language-environment)
297 (assoc current-language-environment
298 language-info-alist))))))
299 (if (or (not mail-parse-mule-charset)
300 (eq mail-parse-mule-charset 'ascii))
301 (setq mail-parse-mule-charset
302 (or (car (last (assq mail-parse-charset
303 mm-mime-mule-charset-alist)))
305 mail-parse-mule-charset)))))))
307 (defun mm-mime-charset (charset)
308 "Return the MIME charset corresponding to the MULE CHARSET."
309 (if (eq charset 'unknown)
310 (error "8-bit characters are found in the message, please specify charset."))
311 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
312 ;; This exists in Emacs 20.
314 (and (mm-preferred-coding-system charset)
316 (mm-preferred-coding-system charset) 'mime-charset))
317 (and (eq charset 'ascii)
319 (mm-preferred-coding-system charset)
320 (mm-mule-charset-to-mime-charset charset))
321 ;; This is for XEmacs.
322 (mm-mule-charset-to-mime-charset charset)))
324 (defun mm-delete-duplicates (list)
325 "Simple substitute for CL `delete-duplicates', testing with `equal'."
328 (setq head (car list))
329 (setq list (delete head list))
330 (setq result (cons head result)))
333 (defun mm-find-mime-charset-region (b e)
334 "Return the MIME charsets needed to encode the region between B and E."
335 (let ((charsets (mapcar 'mm-mime-charset
337 (mm-find-charset-region b e)))))
338 (when (memq 'iso-2022-jp-2 charsets)
339 (setq charsets (delq 'iso-2022-jp charsets)))
340 (setq charsets (mm-delete-duplicates charsets))
341 (if (and (> (length charsets) 1)
342 (fboundp 'find-coding-systems-region)
343 (memq 'utf-8 (find-coding-systems-region b e)))
347 (defsubst mm-multibyte-p ()
348 "Say whether multibyte is enabled."
349 (if (and (not (featurep 'xemacs))
350 (boundp 'enable-multibyte-characters))
351 enable-multibyte-characters
354 (defmacro mm-with-unibyte-buffer (&rest forms)
355 "Create a temporary buffer, and evaluate FORMS there like `progn'.
356 Use unibyte mode for this."
357 `(let (default-enable-multibyte-characters)
358 (with-temp-buffer ,@forms)))
359 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
360 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
362 (defmacro mm-with-unibyte-current-buffer (&rest forms)
363 "Evaluate FORMS with current current buffer temporarily made unibyte.
364 Also bind `default-enable-multibyte-characters' to nil.
365 Equivalent to `progn' in XEmacs"
366 (let ((buffer (make-symbol "buffer")))
367 `(if (and (not (featurep 'xemacs))
368 (boundp 'enable-multibyte-characters)
369 enable-multibyte-characters
370 (fboundp 'set-buffer-multibyte))
371 (let ((,buffer (current-buffer)))
373 (let (default-enable-multibyte-characters)
374 (set-buffer-multibyte nil)
377 (set-buffer-multibyte t)))
378 (let (default-enable-multibyte-characters)
380 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
381 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
383 (defmacro mm-with-unibyte-current-buffer-mule4 (&rest forms)
384 "Evaluate FORMS there like `progn' in current buffer.
386 (let ((buffer (make-symbol "buffer")))
387 `(if (and (not (featurep 'xemacs))
388 (boundp 'enable-multibyte-characters)
389 enable-multibyte-characters
390 (fboundp 'set-buffer-multibyte)
392 (not (charsetp 'eight-bit-control))) ;; For Emacs Mule 4 only.
393 (let ((,buffer (current-buffer)))
395 (let (default-enable-multibyte-characters)
396 (set-buffer-multibyte nil)
399 (set-buffer-multibyte t)))
400 (let (default-enable-multibyte-characters)
402 (put 'mm-with-unibyte-current-buffer-mule4 'lisp-indent-function 0)
403 (put 'mm-with-unibyte-current-buffer-mule4 'edebug-form-spec '(body))
405 (defmacro mm-with-unibyte (&rest forms)
406 "Eval the FORMS with the default value of `enable-multibyte-characters' nil, ."
407 `(let (default-enable-multibyte-characters)
409 (put 'mm-with-unibyte 'lisp-indent-function 0)
410 (put 'mm-with-unibyte 'edebug-form-spec '(body))
412 (defun mm-find-charset-region (b e)
413 "Return a list of Emacs charsets in the region B to E."
415 ((and (mm-multibyte-p)
416 (fboundp 'find-charset-region))
417 ;; Remove composition since the base charsets have been included.
418 ;; Remove eight-bit-*, treat them as ascii.
419 (let ((css (find-charset-region b e)))
420 (mapcar (lambda (cs) (setq css (delq cs css)))
421 '(composition eight-bit-control eight-bit-graphic))
424 ;; We are in a unibyte buffer or XEmacs non-mule, so we futz around a bit.
427 (narrow-to-region b e)
428 (goto-char (point-min))
429 (skip-chars-forward "\0-\177")
434 (and (boundp 'current-language-environment)
435 (car (last (assq 'charset
436 (assoc current-language-environment
437 language-info-alist))))))
438 (if (eq charset 'ascii) (setq charset nil))
441 (car (last (assq mail-parse-charset
442 mm-mime-mule-charset-alist)))))
443 (list 'ascii (or charset 'latin-iso8859-1)))))))))
445 (static-if (fboundp 'shell-quote-argument)
446 (defalias 'mm-quote-arg 'shell-quote-argument)
447 (defun mm-quote-arg (arg)
448 "Return a version of ARG that is safe to evaluate in a shell."
449 (let ((pos 0) new-pos accum)
450 ;; *** bug: we don't handle newline characters properly
451 (while (setq new-pos (string-match "[]*[;!'`\"$\\& \t{} |()<>]" arg pos))
452 (push (substring arg pos new-pos) accum)
454 (push (list (aref arg new-pos)) accum)
455 (setq pos (1+ new-pos)))
458 (apply 'concat (nconc (nreverse accum) (list (substring arg pos))))))))
460 (defun mm-auto-mode-alist ()
461 "Return an `auto-mode-alist' with only the .gz (etc) thingies."
462 (let ((alist auto-mode-alist)
465 (when (listp (cdar alist))
466 (push (car alist) out))
470 (defvar mm-inhibit-file-name-handlers
472 "A list of handlers doing (un)compression (etc) thingies.")
474 (defun mm-insert-file-contents (filename &optional visit beg end replace
476 "Like `insert-file-contents', q.v., but only reads in the file.
477 A buffer may be modified in several ways after reading into the buffer due
478 to advanced Emacs features, such as file-name-handlers, format decoding,
479 find-file-hooks, etc.
480 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers.
481 This function ensures that none of these modifications will take place."
482 (let ((format-alist nil)
483 (auto-mode-alist (if inhibit nil (mm-auto-mode-alist)))
484 (default-major-mode 'fundamental-mode)
485 (enable-local-variables nil)
486 (after-insert-file-functions nil)
487 (enable-local-eval nil)
488 (find-file-hooks nil)
489 (inhibit-file-name-operation (if inhibit
490 'insert-file-contents
491 inhibit-file-name-operation))
492 (inhibit-file-name-handlers
494 (append mm-inhibit-file-name-handlers
495 inhibit-file-name-handlers)
496 inhibit-file-name-handlers)))
497 (insert-file-contents filename visit beg end replace)))
499 (defun mm-append-to-file (start end filename &optional codesys inhibit)
500 "Append the contents of the region to the end of file FILENAME.
501 When called from a function, expects three arguments,
502 START, END and FILENAME. START and END are buffer positions
503 saying what text to write.
504 Optional fourth argument specifies the coding system to use when
506 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
507 (let ((coding-system-for-write
508 (or codesys mm-text-coding-system-for-write
509 mm-text-coding-system))
510 (inhibit-file-name-operation (if inhibit
512 inhibit-file-name-operation))
513 (inhibit-file-name-handlers
515 (append mm-inhibit-file-name-handlers
516 inhibit-file-name-handlers)
517 inhibit-file-name-handlers)))
518 (append-to-file start end filename)))
520 (defun mm-write-region (start end filename &optional append visit lockname
521 coding-system inhibit)
523 "Like `write-region'.
524 If INHIBIT is non-nil, inhibit mm-inhibit-file-name-handlers."
525 (let ((coding-system-for-write
526 (or coding-system mm-text-coding-system-for-write
527 mm-text-coding-system))
528 (inhibit-file-name-operation (if inhibit
530 inhibit-file-name-operation))
531 (inhibit-file-name-handlers
533 (append mm-inhibit-file-name-handlers
534 inhibit-file-name-handlers)
535 inhibit-file-name-handlers)))
536 (write-region start end filename append visit lockname)))
538 (defun mm-image-load-path (&optional package)
540 (dolist (path load-path (nreverse result))
541 (if (file-directory-p
542 (setq dir (concat (file-name-directory
543 (directory-file-name path))
544 "etc/" (or package "gnus/"))))
546 (push path result))))
550 ;;; mm-util.el ends here