1 ;;; utf7.el --- UTF-7 encoding/decoding for Emacs -*-coding: iso-8859-1;-*-
2 ;; Copyright (C) 1999, 2000, 2003 Free Software Foundation, Inc.
4 ;; Author: Jon K Hellan <hellan@acm.org>
5 ;; Maintainer: bugs@gnus.org
8 ;; This file is part of GNU Emacs.
10 ;; GNU Emacs is free software; you can redistribute it and/or modify
11 ;; it under the terms of the GNU General Public License as published by
12 ;; the Free Software Foundation; either version 2, or (at your option)
15 ;; GNU Emacs is distributed in the hope that it will be useful,
16 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
17 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 ;; GNU General Public License for more details.
20 ;; You should have received a copy of the GNU General Public License
21 ;; along with GNU Emacs; see the file COPYING. If not, write to
22 ;; the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
23 ;; Boston, MA 02111-1307, USA.
27 ;; UTF-7 - A Mail-Safe Transformation Format of Unicode - RFC 2152
28 ;; This is a transformation format of Unicode that contains only 7-bit
29 ;; ASCII octets and is intended to be readable by humans in the limiting
30 ;; case that the document consists of characters from the US-ASCII
32 ;; In short, runs of characters outside US-ASCII are encoded as base64
34 ;; A variation of UTF-7 is specified in IMAP 4rev1 (RFC 2060) as the way
35 ;; to represent characters outside US-ASCII in mailbox names in IMAP.
36 ;; This library supports both variants, but the IMAP variation was the
38 ;; The routines convert UTF-7 -> UTF-16 (16 bit encoding of Unicode)
39 ;; -> current character set, and vice versa.
40 ;; However, until Emacs supports Unicode, the only Emacs character set
41 ;; supported here is ISO-8859.1, which can trivially be converted to/from
43 ;; When decoding results in a character outside the Emacs character set,
44 ;; an error is thrown. It is up to the application to recover.
46 ;; UTF-7 should be done by providing a coding system. Mule-UCS does
47 ;; already, but I don't know if it does the IMAP version and it's not
48 ;; clear whether that should really be a coding system. The UTF-16
49 ;; part of the conversion can be done with coding systems available
50 ;; with Mule-UCS or some versions of Emacs. Unfortunately these were
51 ;; done wrongly (regarding handling of byte-order marks and how the
52 ;; variants were named), so we don't have a consistent name for the
53 ;; necessary coding system. The code below doesn't seem to DTRT
56 ;; (utf7-encode "a+£")
59 ;; $ echo "a+£"|iconv -f iso-8859-1 -t utf-7
72 (defconst utf7-direct-encoding-chars " -%'-*,-[]-}"
73 "Character ranges which do not need escaping in UTF-7.")
75 (defconst utf7-imap-direct-encoding-chars
76 (concat utf7-direct-encoding-chars "+\\~")
77 "Character ranges which do not need escaping in the IMAP variant of UTF-7.")
79 (defconst utf7-utf-16-coding-system
80 (cond ((mm-coding-system-p 'utf-16-be-no-signature) ; Mule-UCS
81 'utf-16-be-no-signature)
82 ((and (mm-coding-system-p 'utf-16-be) ; Emacs 21.4 (?), Emacs 22
83 ;; Avoid versions with BOM.
84 (= 2 (length (encode-coding-string "a" 'utf-16-be))))
86 ((mm-coding-system-p 'utf-16-be-nosig) ; ?
88 "Coding system which encodes big endian UTF-16 without a BOM signature.")
90 (defsubst utf7-imap-get-pad-length (len modulus)
91 "Return required length of padding for IMAP modified base64 fragment."
92 (mod (- len) modulus))
94 (defun utf7-encode-internal (&optional for-imap)
95 "Encode text in (temporary) buffer as UTF-7.
96 Use IMAP modification if FOR-IMAP is non-nil."
97 (let ((start (point-min))
99 (narrow-to-region start end)
101 (let* ((esc-char (if for-imap ?& ?+))
102 (direct-encoding-chars
103 (if for-imap utf7-imap-direct-encoding-chars
104 utf7-direct-encoding-chars))
105 (not-direct-encoding-chars (concat "^" direct-encoding-chars)))
107 (skip-chars-forward direct-encoding-chars)
111 (fc (following-char))
113 (skip-chars-forward not-direct-encoding-chars)))
114 (if (and (= fc esc-char)
115 (= run-length 1)) ; Lone esc-char?
116 (delete-backward-char 1) ; Now there's one too many
117 (utf7-fragment-encode p (point) for-imap))
120 (defun utf7-fragment-encode (start end &optional for-imap)
121 "Encode text from START to END in buffer as UTF-7 escape fragment.
122 Use IMAP modification if FOR-IMAP is non-nil."
124 (narrow-to-region start end)
125 (funcall (utf7-get-u16char-converter 'to-utf-16))
126 (mm-with-unibyte-current-buffer
127 (base64-encode-region start (point-max)))
129 (let ((pm (point-max)))
131 (while (search-forward "/" nil t)
132 (replace-match ",")))
133 (skip-chars-forward "^= \t\n" pm)
134 (delete-region (point) pm))))
136 (defun utf7-decode-internal (&optional for-imap)
137 "Decode UTF-7 text in (temporary) buffer.
138 Use IMAP modification if FOR-IMAP is non-nil."
139 (let ((start (point-min))
142 (let* ((esc-pattern (concat "^" (char-to-string (if for-imap ?& ?+))))
143 (base64-chars (concat "A-Za-z0-9+"
144 (char-to-string (if for-imap ?, ?/)))))
146 (skip-chars-forward esc-pattern)
150 (run-length (skip-chars-forward base64-chars)))
151 (when (and (not (eobp)) (= (following-char) ?-))
153 (unless (= run-length 0) ; Encoded lone esc-char?
155 (utf7-fragment-decode p (point) for-imap)
157 (delete-backward-char 1)))))))))
159 (defun utf7-fragment-decode (start end &optional for-imap)
160 "Decode base64 encoded fragment from START to END of UTF-7 text in buffer.
161 Use IMAP modification if FOR-IMAP is non-nil."
163 (narrow-to-region start end)
166 (while (search-forward "," nil 'move-to-end) (replace-match "/")))
167 (let ((pl (utf7-imap-get-pad-length (- end start) 4)))
168 (insert (make-string pl ?=))
169 (base64-decode-region start (+ end pl)))
170 (funcall (utf7-get-u16char-converter 'from-utf-16))))
172 (defun utf7-get-u16char-converter (which-way)
173 "Return a function to convert between UTF-16 and current character set."
174 (if utf7-utf-16-coding-system
175 (if (eq which-way 'to-utf-16)
177 (encode-coding-region (point-min) (point-max)
178 utf7-utf-16-coding-system))
180 (decode-coding-region (point-min) (point-max)
181 utf7-utf-16-coding-system)))
182 ;; Add test to check if we are really Latin-1.
183 (if (eq which-way 'to-utf-16)
184 'utf7-latin1-u16-char-converter
185 'utf7-u16-latin1-char-converter)))
187 (defun utf7-latin1-u16-char-converter ()
188 "Convert latin 1 (ISO-8859.1) characters to 16 bit Unicode.
189 Characters are converted to raw byte pairs in narrowed buffer."
190 (mm-encode-coding-region (point-min) (point-max) 'iso-8859-1)
191 (mm-disable-multibyte)
192 (goto-char (point-min))
197 (defun utf7-u16-latin1-char-converter ()
198 "Convert 16 bit Unicode characters to latin 1 (ISO-8859.1).
199 Characters are in raw byte pairs in narrowed buffer."
200 (goto-char (point-min))
202 (if (= 0 (following-char))
204 (error "Unable to convert from Unicode"))
206 (mm-decode-coding-region (point-min) (point-max) 'iso-8859-1)
207 (mm-enable-multibyte))
209 (defun utf7-encode (string &optional for-imap)
210 "Encode UTF-7 STRING. Use IMAP modification if FOR-IMAP is non-nil."
211 (let ((default-enable-multibyte-characters t))
214 (utf7-encode-internal for-imap)
217 (defun utf7-decode (string &optional for-imap)
218 "Decode UTF-7 STRING. Use IMAP modification if FOR-IMAP is non-nil."
219 (let ((default-enable-multibyte-characters nil))
222 (utf7-decode-internal for-imap)
223 (mm-enable-multibyte)
228 ;;; utf7.el ends here