lisp/gnus/qp.el

   1 ;;; qp.el --- Quoted-Printable functions
   2
   3 ;; Copyright (C) 1998-2011  Free Software Foundation, Inc.
   4
   5 ;; Author: Lars Magne Ingebrigtsen <larsi@gnus.org>
   6 ;; Keywords: mail, extensions
   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 ;; Functions for encoding and decoding quoted-printable text as
  26 ;; defined in RFC 2045.
  27
  28 ;;; Code:
  29
  30 (require 'mm-util)
  31 (defvar mm-use-ultra-safe-encoding)
  32
  33 ;;;###autoload
  34 (defun quoted-printable-decode-region (from to &optional coding-system)
  35   "Decode quoted-printable in the region between FROM and TO, per RFC 2045.
  36 If CODING-SYSTEM is non-nil, decode bytes into characters with that
  37 coding-system.
  38
  39 Interactively, you can supply the CODING-SYSTEM argument
  40 with \\[universal-coding-system-argument].
  41
  42 The CODING-SYSTEM argument is a historical hangover and is deprecated.
  43 QP encodes raw bytes and should be decoded into raw bytes.  Decoding
  44 them into characters should be done separately."
  45   (interactive
  46    ;; Let the user determine the coding system with "C-x RET c".
  47    (list (region-beginning) (region-end) coding-system-for-read))
  48   (unless (mm-coding-system-p coding-system) ; e.g. `ascii' from Gnus
  49     (setq coding-system nil))
  50   (save-excursion
  51     (save-restriction
  52       ;; RFC 2045:  ``An "=" followed by two hexadecimal digits, one
  53       ;; or both of which are lowercase letters in "abcdef", is
  54       ;; formally illegal. A robust implementation might choose to
  55       ;; recognize them as the corresponding uppercase letters.''
  56       (let ((case-fold-search t))
  57         (narrow-to-region from to)
  58         ;; Do this in case we're called from Gnus, say, in a buffer
  59         ;; which already contains non-ASCII characters which would
  60         ;; then get doubly-decoded below.
  61         (if coding-system
  62             (mm-encode-coding-region (point-min) (point-max) coding-system))
  63         (goto-char (point-min))
  64         (while (and (skip-chars-forward "^=")
  65                     (not (eobp)))
  66           (cond ((eq (char-after (1+ (point))) ?\n)
  67                  (delete-char 2))
  68                 ((looking-at "=[0-9A-F][0-9A-F]")
  69                  (let ((byte (string-to-number (buffer-substring (1+ (point))
  70                                                               (+ 3 (point)))
  71                                             16)))
  72                    (mm-insert-byte byte 1)
  73                    (delete-char 3)))
  74                 (t
  75                  (message "Malformed quoted-printable text")
  76                  (forward-char)))))
  77       (if coding-system
  78           (mm-decode-coding-region (point-min) (point-max) coding-system)))))
  79
  80 (defun quoted-printable-decode-string (string &optional coding-system)
  81   "Decode the quoted-printable encoded STRING and return the result.
  82 If CODING-SYSTEM is non-nil, decode the string with coding-system.
  83 Use of CODING-SYSTEM is deprecated; this function should deal with
  84 raw bytes, and coding conversion should be done separately."
  85   (mm-with-unibyte-buffer
  86     (insert string)
  87     (quoted-printable-decode-region (point-min) (point-max) coding-system)
  88     (buffer-string)))
  89
  90 (defun quoted-printable-encode-region (from to &optional fold class)
  91   "Quoted-printable encode the region between FROM and TO per RFC 2045.
  92
  93 If FOLD, fold long lines at 76 characters (as required by the RFC).
  94 If CLASS is non-nil, translate the characters not matched by that
  95 regexp class, which is in the form expected by `skip-chars-forward'.
  96 You should probably avoid non-ASCII characters in this arg.
  97
  98 If `mm-use-ultra-safe-encoding' is set, fold lines unconditionally and
  99 encode lines starting with \"From\"."
 100   (interactive "r")
 101   (unless class
 102     ;; Avoid using 8bit characters. = is \075.
 103     ;; Equivalent to "^\000-\007\013\015-\037\200-\377="
 104     (setq class "\010-\012\014\040-\074\076-\177"))
 105   (save-excursion
 106     (goto-char from)
 107     (if (re-search-forward (mm-string-to-multibyte "[^\x0-\x7f\x80-\xff]")
 108                            to t)
 109         (error "Multibyte character in QP encoding region"))
 110     (save-restriction
 111       (narrow-to-region from to)
 112       ;; Encode all the non-ascii and control characters.
 113       (goto-char (point-min))
 114       (while (and (skip-chars-forward class)
 115                   (not (eobp)))
 116         (insert
 117          (prog1
 118              ;; To unibyte in case of Emacs 23 (unicode) eight-bit.
 119              (format "=%02X" (mm-multibyte-char-to-unibyte (char-after)))
 120            (delete-char 1))))
 121       ;; Encode white space at the end of lines.
 122       (goto-char (point-min))
 123       (while (re-search-forward "[ \t]+$" nil t)
 124         (goto-char (match-beginning 0))
 125         (while (not (eolp))
 126           (insert
 127            (prog1
 128                (format "=%02X" (char-after))
 129              (delete-char 1)))))
 130       (let ((mm-use-ultra-safe-encoding
 131              (and (boundp 'mm-use-ultra-safe-encoding)
 132                   mm-use-ultra-safe-encoding)))
 133         (when (or fold mm-use-ultra-safe-encoding)
 134           (let ((tab-width 1))          ; HTAB is one character.
 135             (goto-char (point-min))
 136             (while (not (eobp))
 137               ;; In ultra-safe mode, encode "From " at the beginning
 138               ;; of a line.
 139               (when mm-use-ultra-safe-encoding
 140                 (if (looking-at "From ")
 141                     (replace-match "From=20" nil t)
 142                   (if (looking-at "-")
 143                       (replace-match "=2D" nil t))))
 144               (end-of-line)
 145               ;; Fold long lines.
 146               (while (> (current-column) 76) ; tab-width must be 1.
 147                 (beginning-of-line)
 148                 (forward-char 75)       ; 75 chars plus an "="
 149                 (search-backward "=" (- (point) 2) t)
 150                 (insert "=\n")
 151                 (end-of-line))
 152               (forward-line))))))))
 153
 154 (defun quoted-printable-encode-string (string)
 155   "Encode the STRING as quoted-printable and return the result."
 156   (with-temp-buffer
 157     (if (mm-multibyte-string-p string)
 158         (mm-enable-multibyte)
 159       (mm-disable-multibyte))
 160     (insert string)
 161     (quoted-printable-encode-region (point-min) (point-max))
 162     (buffer-string)))
 163
 164 (provide 'qp)
 165
 166 ;;; qp.el ends here