1;;; utf-7.el --- utf-7 coding system
2
3;; Copyright (C) 2003, 2004, 2005, 2006, 2007  Free Software Foundation, Inc.
4
5;; Author: Dave Love <fx@gnu.org>
6;; Keywords: i18n, mail
7
8;; This file 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)
11;; any later version.
12
13;; This file 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.
17
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
20;; the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
21;; Boston, MA 02110-1301, USA.
22
23;;; Commentary:
24
25;; Defines a coding system for UTF-7, defined in RFC 2152.  Non-ASCII
26;; segments are encoded as base64-encoded big endian UTF-16.  Also
27;; defines a variation required for IMAP (RFC 2060).
28
29;; The encoding and decoding was originally taken from Jon K Hellan's
30;; implementation in Gnus, but has been substantially re-done.
31
32;; This probably needs more attention.  In particular, it's not
33;; completely consistent with iconv's behaviour.  It's arguable
34;; whether the IMAP version should be a coding system since it's
35;; apparently only used for IMAP mailbox names, so it's commented out.
36
37;;; Code:
38
39;;;###autoload(autoload-coding-system 'utf-7 '(require 'utf-7))
40(make-coding-system
41 'utf-7 0 ?U
42 "UTF-7 encoding of Unicode (RFC 2152)"
43 nil
44 `((safe-chars . ,(coding-system-get 'utf-16be 'safe-chars))
45   (mime-charset . utf-7)
46   (pre-write-conversion . utf-7-pre-write-conversion)
47   (post-read-conversion . utf-7-post-read-conversion)))
48
49;; (make-coding-system
50;;  'utf-7-imap 0 ?u
51;;  "UTF-7 encoding of Unicode, IMAP version (RFC 2060)"
52;;  nil
53;;  `((safe-chars . ,(coding-system-get 'utf-16be 'safe-chars))
54;;    (pre-write-conversion . utf-7-imap-pre-write-conversion)
55;;    (post-read-conversion . utf-7-imap-post-read-conversion)))
56
57(defun utf-7-decode (len imap)
58  "Decode LEN bytes of UTF-7 at point.
59IMAP non-nil means use the IMAP version."
60  (save-excursion
61    (save-restriction
62      (narrow-to-region (point) (+ (point) len))
63      (let ((not-esc (if imap "^&" "^+"))
64	    (skip-chars (if imap "A-Za-z0-9+," "A-Za-z0-9+/")))
65	(while (not (eobp))
66	  (skip-chars-forward not-esc)
67	  (unless (eobp)
68	    (forward-char)
69	    (let ((p (point))
70		  (run-length (skip-chars-forward skip-chars)))
71	      (if (eq ?- (char-after))
72		  (delete-char 1))
73	      (unless (= run-length 0)	; encoded lone esc-char
74		(let ((pl (mod (- run-length) 4)))
75		  (insert-char ?= pl)
76		  (if imap
77		      (subst-char-in-region p (point) ?, ?/))
78		  (base64-decode-region p (point)))
79		(decode-coding-region p (point) 'utf-16be)
80		(save-excursion
81		  (goto-char p)
82		  (delete-backward-char 1)))))))
83      (- (point-max) (point-min)))))
84
85(defun utf-7-post-read-conversion (len)
86  (utf-7-decode len nil))
87
88;; (defun utf-7-imap-post-read-conversion (len)
89;;   (utf-7-decode len t))
90
91(defun utf-7-encode (from to imap)
92  "Encode bytes between FROM and TO to UTF-7.
93ESC and SKIP-CHARS are adjusted for the normal and IMAP versions."
94  (let* ((old-buf (current-buffer))
95	 (esc (if imap ?& ?+))
96	 ;; These are characters which can be encoded asis.
97	 (skip-chars (if imap
98			 "\t\n\r\x20-\x25\x27-\x7e" ; rfc2060
99		       ;; This includes the rfc2152 optional set.
100		       ;; Perhaps it shouldn't (like iconv).
101		       "\t\n\r -*,-[]-}"))
102	 (not-skip-chars (format "^%s%c" skip-chars esc)))
103    (set-buffer (generate-new-buffer " *temp*"))
104    (if (stringp from)
105	(insert from)
106      (insert-buffer-substring old-buf from to))
107    (goto-char (point-min))
108    (while (not (eobp))
109      (skip-chars-forward skip-chars)
110      (if (eq ?+ (char-after))
111	  (progn (forward-char)
112		 (insert ?-))
113	(unless (eobp)
114	  (insert esc)
115	  (let ((p (point)))
116	    (skip-chars-forward not-skip-chars)
117	    (save-restriction
118	      ;; encode-coding-region doesn't preserve point
119	      (narrow-to-region p (point))
120	      (encode-coding-region p (point-max) 'utf-16be)
121	      (base64-encode-region p (point-max))
122	      (if imap
123		  (subst-char-in-region p (point-max) ?/ ?,))
124	      (goto-char p)
125	      ;; As I read the RFC, this isn't correct, but it's
126	      ;; consistent with iconv, at least regarding `='.
127	      (skip-chars-forward "^= \t\n")
128	      (delete-region (point) (point-max))))
129	  (unless (eobp)
130	    (insert ?-)))))
131    nil))
132
133(defun utf-7-pre-write-conversion (from to)
134  (utf-7-encode from to nil))
135
136;; (defun utf-7-imap-pre-write-conversion (from to)
137;;   (utf-7-encode from to t))
138
139(provide 'utf-7)
140
141;;; arch-tag: 975ee403-90a4-4286-97d2-4ed1323f4ef9
142;;; utf-7.el ends here
143