Commit | Line | Data |
---|---|---|
c113de23 GM |
1 | ;;; utf7.el --- UTF-7 encoding/decoding for Emacs |
2 | ;; Copyright (C) 1999, 2000 Free Software Foundation, Inc. | |
3 | ||
4 | ;; Author: Jon K Hellan <hellan@acm.org> | |
5 | ;; Keywords: mail | |
6 | ||
7 | ;; This file is part of GNU Emacs. | |
8 | ||
9 | ;; GNU Emacs is free software; you can redistribute it and/or modify | |
10 | ;; it under the terms of the GNU General Public License as published by | |
11 | ;; the Free Software Foundation; either version 2, or (at your option) | |
12 | ;; any later version. | |
13 | ||
14 | ;; GNU Emacs is distributed in the hope that it will be useful, | |
15 | ;; but WITHOUT ANY WARRANTY; without even the implied warranty of | |
16 | ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the | |
17 | ;; GNU General Public License for more details. | |
18 | ||
19 | ;; You should have received a copy of the GNU General Public License | |
20 | ;; along with GNU Emacs; see the file COPYING. If not, write to | |
21 | ;; the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | |
22 | ;; Boston, MA 02111-1307, USA. | |
23 | ||
24 | ;;; Commentary: | |
25 | ;;; UTF-7 - A Mail-Safe Transformation Format of Unicode - RFC 2152 | |
26 | ;;; This is a transformation format of Unicode that contains only 7-bit | |
27 | ;;; ASCII octets and is intended to be readable by humans in the limiting | |
28 | ;;; case that the document consists of characters from the US-ASCII | |
29 | ;;; repertoire. | |
30 | ;;; In short, runs of characters outside US-ASCII are encoded as base64 | |
31 | ;;; inside delimiters. | |
32 | ;;; A variation of UTF-7 is specified in IMAP 4rev1 (RFC 2060) as the way | |
33 | ;;; to represent characters outside US-ASCII in mailbox names in IMAP. | |
34 | ;;; This library supports both variants, but the IMAP variation was the | |
35 | ;;; reason I wrote it. | |
36 | ;;; The routines convert UTF-7 -> UTF-16 (16 bit encoding of Unicode) | |
37 | ;;; -> current character set, and vice versa. | |
38 | ;;; However, until Emacs supports Unicode, the only Emacs character set | |
39 | ;;; supported here is ISO-8859.1, which can trivially be converted to/from | |
40 | ;;; Unicode. | |
41 | ;;; When decoding results in a character outside the Emacs character set, | |
42 | ;;; an error is thrown. It is up to the application to recover. | |
43 | ||
44 | ;;; Code: | |
45 | ||
46 | (require 'base64) | |
5ce3df57 | 47 | (eval-when-compile (require 'cl)) |
c113de23 GM |
48 | |
49 | (defvar utf7-direct-encoding-chars " -%'-*,-[]-}" | |
50 | "Character ranges which do not need escaping in UTF-7.") | |
51 | ||
52 | (defvar utf7-imap-direct-encoding-chars | |
53 | (concat utf7-direct-encoding-chars "+\\~") | |
54 | "Character ranges which do not need escaping in the IMAP variant of UTF-7.") | |
55 | ||
56 | (defsubst utf7-imap-get-pad-length (len modulus) | |
57 | "Return required length of padding for IMAP modified base64 fragment." | |
58 | (mod (- len) modulus)) | |
59 | ||
60 | (defun utf7-encode-internal (&optional for-imap) | |
61 | "Encode text in (temporary) buffer as UTF-7. | |
62 | Use IMAP modification if FOR-IMAP is non-nil." | |
63 | (let ((start (point-min)) | |
64 | (end (point-max))) | |
65 | (narrow-to-region start end) | |
66 | (goto-char start) | |
67 | (let ((esc-char (if for-imap ?& ?+)) | |
68 | (direct-encoding-chars | |
69 | (if for-imap utf7-imap-direct-encoding-chars | |
70 | utf7-direct-encoding-chars))) | |
71 | (while (not (eobp)) | |
72 | (skip-chars-forward direct-encoding-chars) | |
73 | (unless (eobp) | |
74 | (insert esc-char) | |
75 | (let ((p (point)) | |
76 | (fc (following-char)) | |
77 | (run-length | |
78 | (skip-chars-forward (concat "^" direct-encoding-chars)))) | |
79 | (if (and (= fc esc-char) | |
80 | (= run-length 1)) ; Lone esc-char? | |
81 | (delete-backward-char 1) ; Now there's one too many | |
82 | (utf7-fragment-encode p (point) for-imap)) | |
83 | (insert "-"))))))) | |
84 | ||
85 | (defun utf7-fragment-encode (start end &optional for-imap) | |
86 | "Encode text from START to END in buffer as UTF-7 escape fragment. | |
87 | Use IMAP modification if FOR-IMAP is non-nil." | |
88 | (save-restriction | |
89 | (narrow-to-region start end) | |
90 | (funcall (utf7-get-u16char-converter 'to-utf-16)) | |
91 | (base64-encode-region start (point-max)) | |
92 | (goto-char start) | |
93 | (let ((pm (point-max))) | |
94 | (when for-imap | |
95 | (while (search-forward "/" nil t) | |
96 | (replace-match ","))) | |
97 | (skip-chars-forward "^= \t\n" pm) | |
98 | (delete-region (point) pm)))) | |
99 | ||
100 | (defun utf7-decode-internal (&optional for-imap) | |
101 | "Decode UTF-7 text in (temporary) buffer. | |
102 | Use IMAP modification if FOR-IMAP is non-nil." | |
103 | (let ((start (point-min)) | |
104 | (end (point-max))) | |
105 | (goto-char start) | |
106 | (let* ((esc-pattern (concat "^" (char-to-string (if for-imap ?& ?+)))) | |
107 | (base64-chars (concat "A-Za-z0-9+" | |
108 | (char-to-string (if for-imap ?, ?/))))) | |
109 | (while (not (eobp)) | |
110 | (skip-chars-forward esc-pattern) | |
111 | (unless (eobp) | |
112 | (forward-char) | |
113 | (let ((p (point)) | |
114 | (run-length (skip-chars-forward base64-chars))) | |
115 | (when (and (not (eobp)) (= (following-char) ?-)) | |
116 | (delete-char 1)) | |
117 | (unless (= run-length 0) ; Encoded lone esc-char? | |
118 | (save-excursion | |
119 | (utf7-fragment-decode p (point) for-imap) | |
120 | (goto-char p) | |
121 | (delete-backward-char 1))))))))) | |
122 | ||
123 | (defun utf7-fragment-decode (start end &optional for-imap) | |
124 | "Decode base64 encoded fragment from START to END of UTF-7 text in buffer. | |
125 | Use IMAP modification if FOR-IMAP is non-nil." | |
126 | (save-restriction | |
127 | (narrow-to-region start end) | |
128 | (when for-imap | |
129 | (goto-char start) | |
130 | (while (search-forward "," nil 'move-to-end) (replace-match "/"))) | |
131 | (let ((pl (utf7-imap-get-pad-length (- end start) 4))) | |
132 | (insert (make-string pl ?=)) | |
133 | (base64-decode-region start (+ end pl))) | |
134 | (funcall (utf7-get-u16char-converter 'from-utf-16)))) | |
135 | ||
136 | (defun utf7-get-u16char-converter (which-way) | |
137 | "Return a function to convert between UTF-16 and current character set." | |
138 | ;; Add test to check if we are really Latin-1. | |
139 | ;; Support other character sets once Emacs groks Unicode. | |
140 | (if (eq which-way 'to-utf-16) | |
141 | 'utf7-latin1-u16-char-converter | |
142 | 'utf7-u16-latin1-char-converter)) | |
143 | ||
144 | (defun utf7-latin1-u16-char-converter () | |
145 | "Convert latin 1 (ISO-8859.1) characters to 16 bit Unicode. | |
146 | Characters are converted to raw byte pairs in narrowed buffer." | |
147 | (goto-char (point-min)) | |
148 | (while (not (eobp)) | |
149 | (insert 0) | |
150 | (forward-char))) | |
151 | ||
152 | (defun utf7-u16-latin1-char-converter () | |
153 | "Convert 16 bit Unicode characters to latin 1 (ISO-8859.1). | |
154 | Characters are in raw byte pairs in narrowed buffer." | |
155 | (goto-char (point-min)) | |
156 | (while (not (eobp)) | |
157 | (if (= 0 (following-char)) | |
158 | (delete-char 1) | |
159 | (error "Unable to convert from Unicode")) | |
160 | (forward-char))) | |
161 | ||
162 | (defun utf7-encode (string &optional for-imap) | |
163 | "Encode UTF-7 STRING. Use IMAP modification if FOR-IMAP is non-nil." | |
164 | (let ((default-enable-multibyte-characters nil)) | |
165 | (with-temp-buffer | |
166 | (insert string) | |
167 | (utf7-encode-internal for-imap) | |
168 | (buffer-string)))) | |
169 | ||
170 | (defun utf7-decode (string &optional for-imap) | |
171 | "Decode UTF-7 STRING. Use IMAP modification if FOR-IMAP is non-nil." | |
172 | (let ((default-enable-multibyte-characters nil)) | |
173 | (with-temp-buffer | |
174 | (insert string) | |
175 | (utf7-decode-internal for-imap) | |
176 | (buffer-string)))) | |
177 | ||
178 | (provide 'utf7) | |
179 | ||
ab5796a9 | 180 | ;;; arch-tag: 96078b55-85c7-4161-aed2-932c24b282c7 |
c113de23 | 181 | ;;; utf7.el ends here |