1 ;;; utf-7.el --- utf-7 coding system
3 ;; Copyright (C) 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010
4 ;; Free Software Foundation, Inc.
6 ;; Author: Dave Love <fx@gnu.org>
7 ;; Keywords: i18n, mail
9 ;; This file is part of GNU Emacs.
11 ;; GNU Emacs is free software: you can redistribute it and/or modify
12 ;; it under the terms of the GNU General Public License as published by
13 ;; the Free Software Foundation, either version 3 of the License, or
14 ;; (at your option) any later version.
16 ;; GNU Emacs is distributed in the hope that it will be useful,
17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
19 ;; GNU General Public License for more details.
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with GNU Emacs. If not, see <http://www.gnu.org/licenses/>.
26 ;; Defines a coding system for UTF-7, defined in RFC 2152. Non-ASCII
27 ;; segments are encoded as base64-encoded big endian UTF-16. Also
28 ;; defines a variation required for IMAP (RFC 2060).
30 ;; The encoding and decoding was originally taken from Jon K Hellan's
31 ;; implementation in Gnus, but has been substantially re-done.
33 ;; This probably needs more attention. In particular, it's not
34 ;; completely consistent with iconv's behavior. It's arguable
35 ;; whether the IMAP version should be a coding system since it's
36 ;; apparently only used for IMAP mailbox names, so it's commented out.
40 (defun utf-7-decode (len imap
)
41 "Decode LEN bytes of UTF-7 at point.
42 IMAP non-nil means use the IMAP version."
45 (narrow-to-region (point) (+ (point) len
))
46 (let ((not-esc (if imap
"^&" "^+"))
47 (skip-chars (if imap
"A-Za-z0-9+," "A-Za-z0-9+/")))
49 (skip-chars-forward not-esc
)
53 (run-length (skip-chars-forward skip-chars
)))
54 (if (eq ?-
(char-after))
56 (unless (= run-length
0) ; encoded lone esc-char
57 (let ((pl (mod (- run-length
) 4)))
60 (subst-char-in-region p
(point) ?
, ?
/))
61 (base64-decode-region p
(point)))
62 (decode-coding-region p
(point) 'utf-16be
)
65 (delete-char -
1)))))))
66 (- (point-max) (point-min)))))
69 (defun utf-7-post-read-conversion (len)
70 (utf-7-decode len nil
))
73 (defun utf-7-imap-post-read-conversion (len)
76 (defun utf-7-encode (from to imap
)
77 "Encode bytes between FROM and TO to UTF-7.
78 ESC and SKIP-CHARS are adjusted for the normal and IMAP versions."
79 (let* ((old-buf (current-buffer))
81 ;; These are characters which can be encoded asis.
83 "\t\n\r\x20-\x25\x27-\x7e" ; rfc2060
84 ;; This includes the rfc2152 optional set.
85 ;; Perhaps it shouldn't (like iconv).
87 (not-skip-chars (format "^%s%c" skip-chars esc
)))
88 (set-buffer (generate-new-buffer " *temp*"))
91 (insert-buffer-substring old-buf from to
))
92 (goto-char (point-min))
94 (skip-chars-forward skip-chars
)
95 (if (eq esc
(char-after))
101 (skip-chars-forward not-skip-chars
)
103 ;; encode-coding-region doesn't preserve point
104 (narrow-to-region p
(point))
105 (encode-coding-region p
(point-max) 'utf-16be
)
106 (base64-encode-region p
(point-max))
108 (subst-char-in-region p
(point-max) ?
/ ?
,))
110 ;; As I read the RFC, this isn't correct, but it's
111 ;; consistent with iconv, at least regarding `='.
112 (skip-chars-forward "^= \t\n")
113 (delete-region (point) (point-max))))
114 ;; RFC2060 stipulates that all names MUST end in US-ASCII (i.e.
115 ;; a name that ends with a Unicode octet MUST end with a "-").
116 (if (or imap
(not (eobp)))
121 (defun utf-7-pre-write-conversion (from to
)
122 (utf-7-encode from to nil
))
125 (defun utf-7-imap-pre-write-conversion (from to
)
126 (utf-7-encode from to t
))
130 ;; arch-tag: 975ee403-90a4-4286-97d2-4ed1323f4ef9
131 ;;; utf-7.el ends here