lisp/gnus/rfc1843.el

   1 ;;; rfc1843.el --- HZ (rfc1843) decoding
   2
   3 ;; Copyright (C) 1998, 1999, 2000, 2001, 2002, 2003, 2004,
   4 ;;   2005, 2006, 2007, 2008, 2009, 2010 Free Software Foundation, Inc.
   5
   6 ;; Author: Shenghuo Zhu <zsh@cs.rochester.edu>
   7 ;; Keywords: news HZ HZ+ mail i18n
   8
   9 ;; This file is part of GNU Emacs.
  10
  11 ;; GNU Emacs is free software: you can redistribute it and/or modify
  12 ;; it under the terms of the GNU General Public License as published by
  13 ;; the Free Software Foundation, either version 3 of the License, or
  14 ;; (at your option) any later version.
  15
  16 ;; GNU Emacs is distributed in the hope that it will be useful,
  17 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  19 ;; GNU General Public License for more details.
  20
  21 ;; You should have received a copy of the GNU General Public License
  22 ;; along with GNU Emacs.  If not, see <http://www.gnu.org/licenses/>.
  23
  24 ;;; Commentary:
  25
  26 ;; Usage:
  27 ;; (require 'rfc1843)
  28 ;; (rfc1843-gnus-setup)
  29 ;;
  30 ;; Test:
  31 ;; (rfc1843-decode-string  "~{<:Ky2;S{#,NpJ)l6HK!#~}")
  32
  33 ;;; Code:
  34
  35 ;; For Emacs < 22.2.
  36 (eval-and-compile
  37   (unless (fboundp 'declare-function) (defmacro declare-function (&rest r))))
  38
  39 (eval-when-compile (require 'cl))
  40 (require 'mm-util)
  41
  42 (defvar gnus-decode-encoded-word-function)
  43 (defvar gnus-decode-header-function)
  44 (defvar gnus-newsgroup-name)
  45
  46 (defvar rfc1843-word-regexp
  47   "~\\({\\([\041-\167][\041-\176]\\| \\)+\\)\\(~}\\|$\\)")
  48
  49 (defvar rfc1843-word-regexp-strictly
  50   "~\\({\\([\041-\167][\041-\176]\\)+\\)\\(~}\\|$\\)")
  51
  52 (defvar rfc1843-hzp-word-regexp
  53   "~\\({\\([\041-\167][\041-\176]\\| \\)+\\|\
  54 \[<>]\\([\041-\175][\041-\176]\\| \\)+\\)\\(~}\\|$\\)")
  55
  56 (defvar rfc1843-hzp-word-regexp-strictly
  57   "~\\({\\([\041-\167][\041-\176]\\)+\\|\
  58 \[<>]\\([\041-\175][\041-\176]\\)+\\)\\(~}\\|$\\)")
  59
  60 (defcustom rfc1843-decode-loosely nil
  61   "Loosely check HZ encoding if non-nil.
  62 When it is set non-nil, only buffers or strings with strictly
  63 HZ-encoded are decoded."
  64   :type 'boolean
  65   :group 'mime)
  66
  67 (defcustom rfc1843-decode-hzp t
  68   "HZ+ decoding support if non-nil.
  69 HZ+ specification (also known as HZP) is to provide a standardized
  70 7-bit representation of mixed Big5, GB, and ASCII text for convenient
  71 e-mail transmission, news posting, etc.
  72 The document of HZ+ 0.78 specification can be found at
  73 ftp://ftp.math.psu.edu/pub/simpson/chinese/hzp/hzp.doc"
  74   :type 'boolean
  75   :group 'mime)
  76
  77 (defcustom rfc1843-newsgroups-regexp "chinese\\|hz"
  78   "Regexp of newsgroups in which might be HZ encoded."
  79   :type 'string
  80   :group 'mime)
  81
  82 (defun rfc1843-decode-region (from to)
  83   "Decode HZ in the region between FROM and TO."
  84   (interactive "r")
  85   (let (str firstc)
  86     (save-excursion
  87       (goto-char from)
  88       (if (or rfc1843-decode-loosely
  89               (re-search-forward (if rfc1843-decode-hzp
  90                                      rfc1843-hzp-word-regexp-strictly
  91                                    rfc1843-word-regexp-strictly) to t))
  92           (save-restriction
  93             (narrow-to-region from to)
  94             (goto-char (point-min))
  95             (while (re-search-forward (if rfc1843-decode-hzp
  96                                           rfc1843-hzp-word-regexp
  97                                         rfc1843-word-regexp) (point-max) t)
  98               ;;; Text with extents may cause XEmacs crash
  99               (setq str (buffer-substring-no-properties
 100                          (match-beginning 1)
 101                          (match-end 1)))
 102               (setq firstc (aref str 0))
 103               (insert (mm-decode-coding-string
 104                        (rfc1843-decode
 105                         (prog1
 106                             (substring str 1)
 107                           (delete-region (match-beginning 0) (match-end 0)))
 108                         firstc)
 109                        (if (eq firstc ?{) 'cn-gb-2312 'cn-big5))))
 110             (goto-char (point-min))
 111             (while (search-forward "~" (point-max) t)
 112               (cond ((eq (char-after) ?\n)
 113                      (delete-char -1)
 114                      (delete-char 1))
 115                     ((eq (char-after) ?~)
 116                      (delete-char 1)))))))))
 117
 118 (defun rfc1843-decode-string (string)
 119   "Decode HZ STRING and return the results."
 120   (let ((m (mm-multibyte-p)))
 121     (with-temp-buffer
 122       (when m
 123         (mm-enable-multibyte))
 124       (insert string)
 125       (inline
 126         (rfc1843-decode-region (point-min) (point-max)))
 127       (buffer-string))))
 128
 129 (defun rfc1843-decode (word &optional firstc)
 130   "Decode HZ WORD and return it."
 131   (let ((i -1) (s (substring word 0)) v)
 132     (if (or (not firstc) (eq firstc ?{))
 133         (while (< (incf i) (length s))
 134           (if (eq (setq v (aref s i)) ? ) nil
 135             (aset s i (+ 128 v))))
 136       (while (< (incf i) (length s))
 137         (if (eq (setq v (aref s i)) ? ) nil
 138           (setq v (+ (* 94 v) (aref s (1+ i)) -3135))
 139           (aset s i (+ (/ v 157) (if (eq firstc ?<) 201 161)))
 140           (setq v (% v 157))
 141           (aset s (incf i) (+ v (if (< v 63) 64 98))))))
 142     s))
 143
 144 (autoload 'mail-header-parse-content-type "mail-parse")
 145 (autoload 'message-narrow-to-head "message")
 146 (declare-function message-fetch-field "message" (header &optional not-all))
 147
 148 (defun rfc1843-decode-article-body ()
 149   "Decode HZ encoded text in the article body."
 150   (if (string-match (concat "\\<\\(" rfc1843-newsgroups-regexp "\\)\\>")
 151                     (or gnus-newsgroup-name ""))
 152       (save-excursion
 153         (save-restriction
 154           (message-narrow-to-head)
 155           (let* ((inhibit-point-motion-hooks t)
 156                  (case-fold-search t)
 157                  (ct (message-fetch-field "Content-Type" t))
 158                  (ctl (and ct (mail-header-parse-content-type ct))))
 159             (if (and ctl (not (string-match "/" (car ctl))))
 160                 (setq ctl nil))
 161             (goto-char (point-max))
 162             (widen)
 163             (forward-line 1)
 164             (narrow-to-region (point) (point-max))
 165             (when (or (not ctl)
 166                       (equal (car ctl) "text/plain"))
 167               (rfc1843-decode-region (point) (point-max))))))))
 168
 169 (defvar rfc1843-old-gnus-decode-header-function  nil)
 170 (defvar gnus-decode-header-methods)
 171 (defvar gnus-decode-encoded-word-methods)
 172
 173 (defun rfc1843-gnus-setup ()
 174   "Setup HZ decoding for Gnus."
 175   (require 'gnus-art)
 176   (require 'gnus-sum)
 177   (add-hook 'gnus-article-decode-hook 'rfc1843-decode-article-body t)
 178   (setq gnus-decode-encoded-word-function
 179         'gnus-multi-decode-encoded-word-string
 180         gnus-decode-header-function
 181         'gnus-multi-decode-header
 182         gnus-decode-encoded-word-methods
 183         (nconc gnus-decode-encoded-word-methods
 184                (list
 185                 (cons (concat "\\<\\(" rfc1843-newsgroups-regexp "\\)\\>")
 186                       'rfc1843-decode-string)))
 187         gnus-decode-header-methods
 188         (nconc gnus-decode-header-methods
 189                (list
 190                 (cons (concat "\\<\\(" rfc1843-newsgroups-regexp "\\)\\>")
 191                       'rfc1843-decode-region)))))
 192
 193 (provide 'rfc1843)
 194
 195 ;; arch-tag: 5149c301-a6ca-4731-9c9d-ba616e2cb687
 196 ;;; rfc1843.el ends here