1 ;;; rfc1843.el --- HZ (rfc1843) decoding
3 ;; Copyright (C) 1998, 1999, 2000, 2002, 2003, 2004,
4 ;; 2005 Free Software Foundation, Inc.
6 ;; Author: Shenghuo Zhu <zsh@cs.rochester.edu>
7 ;; Keywords: news HZ HZ+ mail i18n
9 ;; This file is part of GNU Emacs.
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
13 ;; by the Free Software Foundation; either version 2, or (at your
14 ;; option) any later version.
16 ;; GNU Emacs is distributed in the hope that it will be useful, but
17 ;; WITHOUT ANY WARRANTY; without even the implied warranty of
18 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 ;; General Public License for more details.
21 ;; You should have received a copy of the GNU General Public License
22 ;; along with GNU Emacs; see the file COPYING. If not, write to the
23 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
24 ;; Boston, MA 02110-1301, USA.
30 ;; (rfc1843-gnus-setup)
33 ;; (rfc1843-decode-string "~{<:Ky2;S{#,NpJ)l6HK!#~}")
37 (eval-when-compile (require 'cl))
40 (defvar rfc1843-word-regexp
41 "~\\({\\([\041-\167][\041-\176]\\| \\)+\\)\\(~}\\|$\\)")
43 (defvar rfc1843-word-regexp-strictly
44 "~\\({\\([\041-\167][\041-\176]\\)+\\)\\(~}\\|$\\)")
46 (defvar rfc1843-hzp-word-regexp
47 "~\\({\\([\041-\167][\041-\176]\\| \\)+\\|\
48 \[<>]\\([\041-\175][\041-\176]\\| \\)+\\)\\(~}\\|$\\)")
50 (defvar rfc1843-hzp-word-regexp-strictly
51 "~\\({\\([\041-\167][\041-\176]\\)+\\|\
52 \[<>]\\([\041-\175][\041-\176]\\)+\\)\\(~}\\|$\\)")
54 (defcustom rfc1843-decode-loosely nil
55 "Loosely check HZ encoding if non-nil.
56 When it is set non-nil, only buffers or strings with strictly
57 HZ-encoded are decoded."
61 (defcustom rfc1843-decode-hzp t
62 "HZ+ decoding support if non-nil.
63 HZ+ specification (also known as HZP) is to provide a standardized
64 7-bit representation of mixed Big5, GB, and ASCII text for convenient
65 e-mail transmission, news posting, etc.
66 The document of HZ+ 0.78 specification can be found at
67 ftp://ftp.math.psu.edu/pub/simpson/chinese/hzp/hzp.doc"
71 (defcustom rfc1843-newsgroups-regexp "chinese\\|hz"
72 "Regexp of newsgroups in which might be HZ encoded."
76 (defun rfc1843-decode-region (from to)
77 "Decode HZ in the region between FROM and TO."
82 (if (or rfc1843-decode-loosely
83 (re-search-forward (if rfc1843-decode-hzp
84 rfc1843-hzp-word-regexp-strictly
85 rfc1843-word-regexp-strictly) to t))
87 (narrow-to-region from to)
88 (goto-char (point-min))
89 (while (re-search-forward (if rfc1843-decode-hzp
90 rfc1843-hzp-word-regexp
91 rfc1843-word-regexp) (point-max) t)
92 ;;; Text with extents may cause XEmacs crash
93 (setq str (buffer-substring-no-properties
96 (setq firstc (aref str 0))
97 (insert (mm-decode-coding-string
101 (delete-region (match-beginning 0) (match-end 0)))
103 (if (eq firstc ?{) 'cn-gb-2312 'cn-big5))))
104 (goto-char (point-min))
105 (while (search-forward "~" (point-max) t)
106 (cond ((eq (char-after) ?\n)
109 ((eq (char-after) ?~)
110 (delete-char 1)))))))))
112 (defun rfc1843-decode-string (string)
113 "Decode HZ STRING and return the results."
114 (let ((m (mm-multibyte-p)))
117 (mm-enable-multibyte))
120 (rfc1843-decode-region (point-min) (point-max)))
123 (defun rfc1843-decode (word &optional firstc)
124 "Decode HZ WORD and return it."
125 (let ((i -1) (s (substring word 0)) v)
126 (if (or (not firstc) (eq firstc ?{))
127 (while (< (incf i) (length s))
128 (if (eq (setq v (aref s i)) ? ) nil
129 (aset s i (+ 128 v))))
130 (while (< (incf i) (length s))
131 (if (eq (setq v (aref s i)) ? ) nil
132 (setq v (+ (* 94 v) (aref s (1+ i)) -3135))
133 (aset s i (+ (/ v 157) (if (eq firstc ?<) 201 161)))
135 (aset s (incf i) (+ v (if (< v 63) 64 98))))))
138 (defun rfc1843-decode-article-body ()
139 "Decode HZ encoded text in the article body."
140 (if (string-match (concat "\\<\\(" rfc1843-newsgroups-regexp "\\)\\>")
141 (or gnus-newsgroup-name ""))
144 (message-narrow-to-head)
145 (let* ((inhibit-point-motion-hooks t)
147 (ct (message-fetch-field "Content-Type" t))
148 (ctl (and ct (ignore-errors
149 (mail-header-parse-content-type ct)))))
150 (if (and ctl (not (string-match "/" (car ctl))))
152 (goto-char (point-max))
155 (narrow-to-region (point) (point-max))
157 (equal (car ctl) "text/plain"))
158 (rfc1843-decode-region (point) (point-max))))))))
160 (defvar rfc1843-old-gnus-decode-header-function nil)
161 (defvar gnus-decode-header-methods)
162 (defvar gnus-decode-encoded-word-methods)
164 (defun rfc1843-gnus-setup ()
165 "Setup HZ decoding for Gnus."
168 (add-hook 'gnus-article-decode-hook 'rfc1843-decode-article-body t)
169 (setq gnus-decode-encoded-word-function
170 'gnus-multi-decode-encoded-word-string
171 gnus-decode-header-function
172 'gnus-multi-decode-header
173 gnus-decode-encoded-word-methods
174 (nconc gnus-decode-encoded-word-methods
176 (cons (concat "\\<\\(" rfc1843-newsgroups-regexp "\\)\\>")
177 'rfc1843-decode-string)))
178 gnus-decode-header-methods
179 (nconc gnus-decode-header-methods
181 (cons (concat "\\<\\(" rfc1843-newsgroups-regexp "\\)\\>")
182 'rfc1843-decode-region)))))
186 ;;; arch-tag: 5149c301-a6ca-4731-9c9d-ba616e2cb687
187 ;;; rfc1843.el ends here