lisp/gnus/utf7.el

   1 ;;; utf7.el --- UTF-7 encoding/decoding for Emacs   -*-coding: iso-8859-1;-*-
   2
   3 ;; Copyright (C) 1999, 2000, 2001, 2002, 2003, 2004,
   4 ;;   2005, 2006, 2007, 2008 Free Software Foundation, Inc.
   5
   6 ;; Author: Jon K Hellan <hellan@acm.org>
   7 ;; Maintainer: bugs@gnus.org
   8 ;; Keywords: mail
   9
  10 ;; This file is part of GNU Emacs.
  11
  12 ;; GNU Emacs is free software: you can redistribute it and/or modify
  13 ;; it under the terms of the GNU General Public License as published by
  14 ;; the Free Software Foundation, either version 3 of the License, or
  15 ;; (at your option) any later version.
  16
  17 ;; GNU Emacs is distributed in the hope that it will be useful,
  18 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  19 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  20 ;; GNU General Public License for more details.
  21
  22 ;; You should have received a copy of the GNU General Public License
  23 ;; along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.
  24
  25 ;;; Commentary:
  26
  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
  31 ;; repertoire.
  32 ;; In short, runs of characters outside US-ASCII are encoded as base64
  33 ;; inside delimiters.
  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
  37 ;; reason I wrote it.
  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
  42 ;; Unicode.
  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.
  45
  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
  54 ;; generally.  E.g.:
  55 ;;
  56 ;; (utf7-encode "a+£")
  57 ;;   => "a+ACsAow-"
  58 ;;
  59 ;; $ echo "a+£"|iconv -f iso-8859-1 -t utf-7
  60 ;; a+-+AKM
  61 ;;
  62 ;;  -- fx
  63
  64
  65 ;;; Code:
  66
  67 (require 'base64)
  68 (eval-when-compile (require 'cl))
  69 (require 'mm-util)
  70
  71 (defconst utf7-direct-encoding-chars " -%'-*,-[]-}"
  72   "Character ranges which do not need escaping in UTF-7.")
  73
  74 (defconst utf7-imap-direct-encoding-chars
  75   (concat utf7-direct-encoding-chars "+\\~")
  76   "Character ranges which do not need escaping in the IMAP variant of UTF-7.")
  77
  78 (defconst utf7-utf-16-coding-system
  79   (cond ((mm-coding-system-p 'utf-16-be-no-signature) ; Mule-UCS
  80          'utf-16-be-no-signature)
  81         ((and (mm-coding-system-p 'utf-16-be) ; Emacs 21.3, Emacs 22
  82               ;; Avoid versions with BOM.
  83               (= 2 (length (encode-coding-string "a" 'utf-16-be))))
  84          'utf-16-be)
  85         ((mm-coding-system-p 'utf-16-be-nosig) ; ?
  86          'utf-16-be-nosig))
  87   "Coding system which encodes big endian UTF-16 without a BOM signature.")
  88
  89 (defsubst utf7-imap-get-pad-length (len modulus)
  90   "Return required length of padding for IMAP modified base64 fragment."
  91   (mod (- len) modulus))
  92
  93 (defun utf7-encode-internal (&optional for-imap)
  94   "Encode text in (temporary) buffer as UTF-7.
  95 Use IMAP modification if FOR-IMAP is non-nil."
  96   (let ((start (point-min))
  97         (end (point-max)))
  98     (narrow-to-region start end)
  99     (goto-char start)
 100     (let* ((esc-char (if for-imap ?& ?+))
 101            (direct-encoding-chars
 102             (if for-imap utf7-imap-direct-encoding-chars
 103               utf7-direct-encoding-chars))
 104            (not-direct-encoding-chars (concat "^" direct-encoding-chars)))
 105       (while (not (eobp))
 106         (skip-chars-forward direct-encoding-chars)
 107         (unless (eobp)
 108           (insert esc-char)
 109           (let ((p (point))
 110                 (fc (following-char))
 111                 (run-length
 112                  (skip-chars-forward not-direct-encoding-chars)))
 113             (if (and (= fc esc-char)
 114                      (= run-length 1))  ; Lone esc-char?
 115                 (delete-backward-char 1) ; Now there's one too many
 116               (utf7-fragment-encode p (point) for-imap))
 117             (insert "-")))))))
 118
 119 (defun utf7-fragment-encode (start end &optional for-imap)
 120   "Encode text from START to END in buffer as UTF-7 escape fragment.
 121 Use IMAP modification if FOR-IMAP is non-nil."
 122   (save-restriction
 123     (narrow-to-region start end)
 124     (funcall (utf7-get-u16char-converter 'to-utf-16))
 125     (mm-with-unibyte-current-buffer
 126       (base64-encode-region start (point-max)))
 127     (goto-char start)
 128     (let ((pm (point-max)))
 129       (when for-imap
 130         (while (search-forward "/" nil t)
 131           (replace-match ",")))
 132       (skip-chars-forward "^= \t\n" pm)
 133       (delete-region (point) pm))))
 134
 135 (defun utf7-decode-internal (&optional for-imap)
 136   "Decode UTF-7 text in (temporary) buffer.
 137 Use IMAP modification if FOR-IMAP is non-nil."
 138   (let ((start (point-min))
 139         (end (point-max)))
 140     (goto-char start)
 141     (let* ((esc-pattern (concat "^" (char-to-string (if for-imap ?& ?+))))
 142            (base64-chars (concat "A-Za-z0-9+"
 143                                  (char-to-string (if for-imap ?, ?/)))))
 144       (while (not (eobp))
 145         (skip-chars-forward esc-pattern)
 146         (unless (eobp)
 147           (forward-char)
 148           (let ((p (point))
 149                 (run-length (skip-chars-forward base64-chars)))
 150             (when (and (not (eobp)) (= (following-char) ?-))
 151               (delete-char 1))
 152             (unless (= run-length 0)    ; Encoded lone esc-char?
 153               (save-excursion
 154                 (utf7-fragment-decode p (point) for-imap)
 155                 (goto-char p)
 156                 (delete-backward-char 1)))))))))
 157
 158 (defun utf7-fragment-decode (start end &optional for-imap)
 159   "Decode base64 encoded fragment from START to END of UTF-7 text in buffer.
 160 Use IMAP modification if FOR-IMAP is non-nil."
 161   (save-restriction
 162     (narrow-to-region start end)
 163     (when for-imap
 164       (goto-char start)
 165       (while (search-forward "," nil 'move-to-end) (replace-match "/")))
 166     (let ((pl (utf7-imap-get-pad-length (- end start) 4)))
 167       (insert (make-string pl ?=))
 168       (base64-decode-region start (+ end pl)))
 169     (funcall (utf7-get-u16char-converter 'from-utf-16))))
 170
 171 (defun utf7-get-u16char-converter (which-way)
 172   "Return a function to convert between UTF-16 and current character set."
 173   (if utf7-utf-16-coding-system
 174       (if (eq which-way 'to-utf-16)
 175           (lambda ()
 176             (encode-coding-region (point-min) (point-max)
 177                                   utf7-utf-16-coding-system))
 178         (lambda ()
 179           (decode-coding-region (point-min) (point-max)
 180                                 utf7-utf-16-coding-system)))
 181     ;; Add test to check if we are really Latin-1.
 182     (if (eq which-way 'to-utf-16)
 183         'utf7-latin1-u16-char-converter
 184       'utf7-u16-latin1-char-converter)))
 185
 186 (defun utf7-latin1-u16-char-converter ()
 187   "Convert latin 1 (ISO-8859.1) characters to 16 bit Unicode.
 188 Characters are converted to raw byte pairs in narrowed buffer."
 189   (mm-encode-coding-region (point-min) (point-max) 'iso-8859-1)
 190   (mm-disable-multibyte)
 191   (goto-char (point-min))
 192   (while (not (eobp))
 193     (insert 0)
 194     (forward-char)))
 195
 196 (defun utf7-u16-latin1-char-converter ()
 197   "Convert 16 bit Unicode characters to latin 1 (ISO-8859.1).
 198 Characters are in raw byte pairs in narrowed buffer."
 199   (goto-char (point-min))
 200   (while (not (eobp))
 201     (if (= 0 (following-char))
 202         (delete-char 1)
 203         (error "Unable to convert from Unicode"))
 204     (forward-char))
 205   (mm-decode-coding-region (point-min) (point-max) 'iso-8859-1)
 206   (mm-enable-multibyte))
 207
 208 (defun utf7-encode (string &optional for-imap)
 209   "Encode UTF-7 STRING.  Use IMAP modification if FOR-IMAP is non-nil."
 210   (if (and (coding-system-p 'utf-7) (coding-system-p 'utf-7-imap))
 211       ;; Emacs 23 with proper support for IMAP
 212       (encode-coding-string string (if for-imap 'utf-7-imap 'utf-7))
 213     (let ((default-enable-multibyte-characters t))
 214       (with-temp-buffer
 215         (insert string)
 216         (utf7-encode-internal for-imap)
 217         (buffer-string)))))
 218
 219 (defun utf7-decode (string &optional for-imap)
 220   "Decode UTF-7 STRING.  Use IMAP modification if FOR-IMAP is non-nil."
 221   (if (and (coding-system-p 'utf-7) (coding-system-p 'utf-7-imap))
 222       ;; Emacs 23 with proper support for IMAP
 223       (decode-coding-string string (if for-imap 'utf-7-imap 'utf-7))
 224     (let ((default-enable-multibyte-characters nil))
 225       (with-temp-buffer
 226         (insert string)
 227         (utf7-decode-internal for-imap)
 228         (mm-enable-multibyte)
 229         (buffer-string)))))
 230
 231 (provide 'utf7)
 232
 233 ;; arch-tag: 96078b55-85c7-4161-aed2-932c24b282c7
 234 ;;; utf7.el ends here