lisp/international/utf-7.el

   1 ;;; utf-7.el --- utf-7 coding system
   2
   3 ;; Copyright (C) 2003-2015 Free Software Foundation, Inc.
   4
   5 ;; Author: Dave Love <fx@gnu.org>
   6 ;; Keywords: i18n, mail
   7
   8 ;; This file is part of GNU Emacs.
   9
  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 3 of the License, or
  13 ;; (at your option) any later version.
  14
  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.
  19
  20 ;; You should have received a copy of the GNU General Public License
  21 ;; along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.
  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 behavior.  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 (defun utf-7-decode (len imap)
  40   "Decode LEN bytes of UTF-7 at point.
  41 IMAP non-nil means use the IMAP version."
  42   (save-excursion
  43     (save-restriction
  44       (narrow-to-region (point) (+ (point) len))
  45       (let ((not-esc (if imap "^&" "^+"))
  46             (skip-chars (if imap "A-Za-z0-9+," "A-Za-z0-9+/")))
  47         (while (not (eobp))
  48           (skip-chars-forward not-esc)
  49           (unless (eobp)
  50             (forward-char)
  51             (let ((p (point))
  52                   (run-length (skip-chars-forward skip-chars)))
  53               (if (eq ?- (char-after))
  54                   (delete-char 1))
  55               (unless (= run-length 0)  ; encoded lone esc-char
  56                 (let ((pl (mod (- run-length) 4)))
  57                   (insert-char ?= pl)
  58                   (if imap
  59                       (subst-char-in-region p (point) ?, ?/))
  60                   (base64-decode-region p (point)))
  61                 (decode-coding-region p (point) 'utf-16be)
  62                 (save-excursion
  63                   (goto-char p)
  64                   (delete-char -1)))))))
  65       (- (point-max) (point-min)))))
  66
  67 ;;;###autoload
  68 (defun utf-7-post-read-conversion (len)
  69   (utf-7-decode len nil))
  70
  71 ;;;###autoload
  72 (defun utf-7-imap-post-read-conversion (len)
  73   (utf-7-decode len t))
  74
  75 (defun utf-7-encode (from to imap)
  76   "Encode bytes between FROM and TO to UTF-7.
  77 ESC and SKIP-CHARS are adjusted for the normal and IMAP versions."
  78   (let* ((old-buf (current-buffer))
  79          (esc (if imap ?& ?+))
  80          ;; These are characters which can be encoded asis.
  81          (skip-chars (if imap
  82                          "\t\n\r\x20-\x25\x27-\x7e" ; rfc2060
  83                        ;; This includes the rfc2152 optional set.
  84                        ;; Perhaps it shouldn't (like iconv).
  85                        "\t\n\r -*,-[]-}"))
  86          (not-skip-chars (format "^%s%c" skip-chars esc)))
  87     (set-buffer (generate-new-buffer " *temp*"))
  88     (if (stringp from)
  89         (insert from)
  90       (insert-buffer-substring old-buf from to))
  91     (goto-char (point-min))
  92     (while (not (eobp))
  93       (skip-chars-forward skip-chars)
  94       (if (eq esc (char-after))
  95           (progn (forward-char)
  96                  (insert ?-))
  97         (unless (eobp)
  98           (insert esc)
  99           (let ((p (point)))
 100             (skip-chars-forward not-skip-chars)
 101             (save-restriction
 102               ;; encode-coding-region doesn't preserve point
 103               (narrow-to-region p (point))
 104               (encode-coding-region p (point-max) 'utf-16be)
 105               (base64-encode-region p (point-max))
 106               (if imap
 107                   (subst-char-in-region p (point-max) ?/ ?,))
 108               (goto-char p)
 109               ;; As I read the RFC, this isn't correct, but it's
 110               ;; consistent with iconv, at least regarding `='.
 111               (skip-chars-forward "^= \t\n")
 112               (delete-region (point) (point-max))))
 113           ;; RFC2060 stipulates that all names MUST end in US-ASCII (i.e.
 114           ;; a name that ends with a Unicode octet MUST end with a "-").
 115           (if (or imap (not (eobp)))
 116             (insert ?-)))))
 117     nil))
 118
 119 ;;;###autoload
 120 (defun utf-7-pre-write-conversion (from to)
 121   (utf-7-encode from to nil))
 122
 123 ;;;###autoload
 124 (defun utf-7-imap-pre-write-conversion (from to)
 125   (utf-7-encode from to t))
 126
 127 (provide 'utf-7)
 128
 129 ;;; utf-7.el ends here