1 ;;; mm-util.el --- Utility functions for Mule and low level 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))
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))
129 (defalias 'mm-char-or-char-int-p
131 ((fboundp 'char-or-char-int-p) 'char-or-char-int-p)
132 ((fboundp 'char-valid-p) 'char-valid-p)
135 (defvar mm-coding-system-list nil)
136 (defun mm-get-coding-system-list ()
137 "Get the coding system list."
138 (or mm-coding-system-list
139 (setq mm-coding-system-list (mm-coding-system-list))))
141 (defun mm-coding-system-p (sym)
142 "Return non-nil if SYM is a coding system."
143 (or (and (fboundp 'coding-system-p) (coding-system-p sym))
144 (memq sym (mm-get-coding-system-list))))
146 (defvar mm-charset-synonym-alist
148 (gb2312 . cn-gb-2312)
150 ;; Windows-1252 is actually a superset of Latin-1. See also
151 ;; `gnus-article-dumbquotes-map'.
152 ,(unless (mm-coding-system-p 'windows-1252) ; should be defined eventually
153 '(windows-1252 . iso-8859-1))
155 "A mapping from invalid charset names to the real charset names.")
157 (defvar mm-binary-coding-system
159 ((mm-coding-system-p 'binary) 'binary)
160 ((mm-coding-system-p 'no-conversion) 'no-conversion)
162 "100% binary coding system.")
164 (defvar mm-text-coding-system
165 (or (if (memq system-type '(windows-nt ms-dos ms-windows))
166 (and (mm-coding-system-p 'raw-text-dos) 'raw-text-dos)
167 (and (mm-coding-system-p 'raw-text) 'raw-text))
168 mm-binary-coding-system)
169 "Text-safe coding system (For removing ^M).")
171 (defvar mm-text-coding-system-for-write nil
172 "Text coding system for write.")
174 (defvar mm-auto-save-coding-system
176 ((mm-coding-system-p 'emacs-mule)
177 (if (memq system-type '(windows-nt ms-dos ms-windows))
178 (if (mm-coding-system-p 'emacs-mule-dos)
179 'emacs-mule-dos mm-binary-coding-system)
181 ((mm-coding-system-p 'escape-quoted) 'escape-quoted)
182 (t mm-binary-coding-system))
183 "Coding system of auto save file.")
185 ;;; Internal variables:
189 (defun mm-mule-charset-to-mime-charset (charset)
190 "Return the MIME charset corresponding to the given Mule CHARSET."
191 (let ((alist mm-mime-mule-charset-alist)
194 (when (memq charset (cdar alist))
195 (setq out (caar alist)
200 (defun mm-charset-to-coding-system (charset &optional lbt)
201 "Return coding-system corresponding to CHARSET.
202 CHARSET is a symbol naming a MIME charset.
203 If optional argument LBT (`unix', `dos' or `mac') is specified, it is
204 used as the line break code type of the coding system."
205 (when (stringp charset)
206 (setq charset (intern (downcase charset))))
208 (or (cdr (assq charset mm-charset-synonym-alist))
211 (setq charset (intern (format "%s-%s" charset lbt))))
213 ;; Running in a non-MULE environment.
214 ((null (mm-get-coding-system-list))
217 ((eq charset 'us-ascii)
219 ;; Check to see whether we can handle this charset. (This depends
220 ;; on there being some coding system matching each `mime-charset'
221 ;; coding sysytem property defined, as there should be.)
222 ((memq charset (mm-get-coding-system-list))
228 (defsubst mm-replace-chars-in-string (string from to)
229 (mm-subst-char-in-string from to string))
231 (defsubst mm-enable-multibyte ()
232 "Set the multibyte flag of the current buffer.
233 Only do this if the default value of `enable-multibyte-characters' is
234 non-nil. This is a no-op in XEmacs."
235 (when (and (not (featurep 'xemacs))
236 (boundp 'default-enable-multibyte-characters)
237 default-enable-multibyte-characters
238 (fboundp 'set-buffer-multibyte))
239 (set-buffer-multibyte t)))
241 (defsubst mm-disable-multibyte ()
242 "Unset the multibyte flag of in the current buffer.
243 This is a no-op in XEmacs."
244 (when (and (not (featurep 'xemacs))
245 (fboundp 'set-buffer-multibyte))
246 (set-buffer-multibyte nil)))
248 (defsubst mm-enable-multibyte-mule4 ()
249 "Enable multibyte in the current buffer.
250 Only used in Emacs Mule 4."
251 (when (and (not (featurep 'xemacs))
252 (boundp 'default-enable-multibyte-characters)
253 default-enable-multibyte-characters
254 (fboundp 'set-buffer-multibyte)
256 (not (charsetp 'eight-bit-control)))
257 (set-buffer-multibyte t)))
259 (defsubst mm-disable-multibyte-mule4 ()
260 "Disable multibyte in the current buffer.
261 Only used in Emacs Mule 4."
262 (when (and (not (featurep 'xemacs))
263 (fboundp 'set-buffer-multibyte)
265 (not (charsetp 'eight-bit-control)))
266 (set-buffer-multibyte nil)))
268 (defun mm-preferred-coding-system (charset)
269 ;; A typo in some Emacs versions.
270 (or (get-charset-property charset 'prefered-coding-system)
271 (get-charset-property charset 'preferred-coding-system)))
273 (defun mm-charset-after (&optional pos)
274 "Return charset of a character in current buffer at position POS.
275 If POS is nil, it defauls to the current point.
276 If POS is out of range, the value is nil.
277 If the charset is `composition', return the actual one."
278 (let ((char (char-after pos)) charset)
279 (if (< (mm-char-int char) 128)
280 (setq charset 'ascii)
281 ;; charset-after is fake in some Emacsen.
282 (setq charset (and (fboundp 'char-charset) (char-charset char)))
283 (if (eq charset 'composition)
284 (let ((p (or pos (point))))
285 (cadr (find-charset-region p (1+ p))))
286 (if (and charset (not (memq charset '(ascii eight-bit-control
287 eight-bit-graphic))))
290 mail-parse-mule-charset ;; cached mule-charset
292 (setq mail-parse-mule-charset
293 (and (boundp 'current-language-environment)
296 (assoc current-language-environment
297 language-info-alist))))))
298 (if (or (not mail-parse-mule-charset)
299 (eq mail-parse-mule-charset 'ascii))
300 (setq mail-parse-mule-charset
301 (or (car (last (assq mail-parse-charset
302 mm-mime-mule-charset-alist)))
304 mail-parse-mule-charset)))))))
306 (defun mm-mime-charset (charset)
307 "Return the MIME charset corresponding to the MULE CHARSET."
308 (if (eq charset 'unknown)
309 (error "8-bit characters are found in the message, please specify charset."))
310 (if (and (fboundp 'coding-system-get) (fboundp 'get-charset-property))
311 ;; This exists in Emacs 20.
313 (and (mm-preferred-coding-system charset)
315 (mm-preferred-coding-system charset) 'mime-charset))
316 (and (eq charset 'ascii)
318 (mm-preferred-coding-system charset)
319 (mm-mule-charset-to-mime-charset charset))
320 ;; This is for XEmacs.
321 (mm-mule-charset-to-mime-charset charset)))
323 (defun mm-delete-duplicates (list)
324 "Simple substitute for CL `delete-duplicates', testing with `equal'."
327 (setq head (car list))
328 (setq list (delete head list))
329 (setq result (cons head result)))
332 (defun mm-find-mime-charset-region (b e)
333 "Return the MIME charsets needed to encode the region between B and E."
334 (let ((charsets (mapcar 'mm-mime-charset
336 (mm-find-charset-region b e)))))
337 (when (memq 'iso-2022-jp-2 charsets)
338 (setq charsets (delq 'iso-2022-jp charsets)))
339 (setq charsets (mm-delete-duplicates charsets))
340 (if (and (> (length charsets) 1)
341 (fboundp 'find-coding-systems-region)
342 (memq 'utf-8 (find-coding-systems-region b e)))
346 (defsubst mm-multibyte-p ()
347 "Say whether multibyte is enabled."
348 (if (and (not (featurep 'xemacs))
349 (boundp 'enable-multibyte-characters))
350 enable-multibyte-characters
353 (defmacro mm-with-unibyte-buffer (&rest forms)
354 "Create a temporary buffer, and evaluate FORMS there like `progn'.
355 Use unibyte mode for this."
356 `(let (default-enable-multibyte-characters)
357 (with-temp-buffer ,@forms)))
358 (put 'mm-with-unibyte-buffer 'lisp-indent-function 0)
359 (put 'mm-with-unibyte-buffer 'edebug-form-spec '(body))
361 (defmacro mm-with-unibyte-current-buffer (&rest forms)
362 "Evaluate FORMS with current current buffer temporarily made unibyte.
363 Also bind `default-enable-multibyte-characters' to nil.
364 Equivalent to `progn' in XEmacs"
365 (let ((buffer (make-symbol "buffer")))
366 `(if (and (not (featurep 'xemacs))
367 (boundp 'enable-multibyte-characters)
368 enable-multibyte-characters
369 (fboundp 'set-buffer-multibyte))
370 (let ((,buffer (current-buffer)))
372 (let (default-enable-multibyte-characters)
373 (set-buffer-multibyte nil)
376 (set-buffer-multibyte t)))
377 (let (default-enable-multibyte-characters)
379 (put 'mm-with-unibyte-current-buffer 'lisp-indent-function 0)
380 (put 'mm-with-unibyte-current-buffer 'edebug-form-spec '(body))
382 (defmacro mm-with-unibyte-current-buffer-mule4 (&rest forms)
383 "Evaluate FORMS there like `progn' in current buffer.
385 (let ((buffer (make-symbol "buffer")))
386 `(if (and (not (featurep 'xemacs))
387 (boundp 'enable-multibyte-characters)
388 enable-multibyte-characters
389 (fboundp 'set-buffer-multibyte)
391 (not (charsetp 'eight-bit-control))) ;; For Emacs Mule 4 only.
392 (let ((,buffer (current-buffer)))
394 (let (default-enable-multibyte-characters)
395 (set-buffer-multibyte nil)
398 (set-buffer-multibyte t)))
399 (let (default-enable-multibyte-characters)
401 (put 'mm-with-unibyte-current-buffer-mule4 'lisp-indent-function 0)
402 (put 'mm-with-unibyte-current-buffer-mule4 'edebug-form-spec '(body))
404 (defmacro mm-with-unibyte (&rest forms)
405 "Eval the FORMS with the default value of `enable-multibyte-characters' nil, ."
406 `(let (default-enable-multibyte-characters)
408 (put 'mm-with-unibyte 'lisp-indent-function 0)
409 (put 'mm-with-unibyte 'edebug-form-spec '(body))
411 (defun mm-find-charset-region (b e)
412 "Return a list of Emacs charsets in the region B to E."
414 ((and (mm-multibyte-p)
415 (fboundp 'find-charset-region))
416 ;; Remove composition since the base charsets have been included.
417 ;; Remove eight-bit-*, treat them as ascii.
418 (let ((css (find-charset-region b e)))
419 (mapcar (lambda (cs) (setq css (delq cs css)))
420 '(composition eight-bit-control eight-bit-graphic))
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 (static-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)))
537 (defun mm-image-load-path (&optional package)
539 (dolist (path load-path (nreverse result))
540 (if (file-directory-p
541 (setq dir (concat (file-name-directory
542 (directory-file-name path))
543 "etc/" (or package "gnus/"))))
545 (push path result))))
549 ;;; mm-util.el ends here