annotate lisp/gnus/html2text.el @ 78192:19db593f63bb

(set-default-process-coding-system): Use dos line ends for input to cmdproxy on all versions of Windows. Use dos line ends for input to plink.
author Jason Rumney <jasonr@gnu.org>
date Sun, 22 Jul 2007 22:53:48 +0000
parents e3694f1cb928
children 24202b793a08
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
1 ;;; html2text.el --- a simple html to plain text converter
64754
fafd692d1e40 Update years in copyright notice; nfc.
Thien-Thi Nguyen <ttn@gnuvola.org>
parents: 64085
diff changeset
2
75347
e3694f1cb928 Add 2007 to copyright years.
Glenn Morris <rgm@gnu.org>
parents: 73627
diff changeset
3 ;; Copyright (C) 2002, 2003, 2004, 2005, 2006, 2007 Free Software Foundation, Inc.
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
4
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
5 ;; Author: Joakim Hove <hove@phys.ntnu.no>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
6
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
7 ;; This file is part of GNU Emacs.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
8
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
9 ;; GNU Emacs is free software; you can redistribute it and/or modify
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
10 ;; it under the terms of the GNU General Public License as published by
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
11 ;; the Free Software Foundation; either version 2, or (at your option)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
12 ;; any later version.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
13
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
14 ;; GNU Emacs is distributed in the hope that it will be useful,
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
15 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
16 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
17 ;; GNU General Public License for more details.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
18
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
19 ;; You should have received a copy of the GNU General Public License
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
20 ;; along with GNU Emacs; see the file COPYING. If not, write to the
64085
18a818a2ee7c Update FSF's address.
Lute Kamstra <lute@gnu.org>
parents: 62943
diff changeset
21 ;; Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
18a818a2ee7c Update FSF's address.
Lute Kamstra <lute@gnu.org>
parents: 62943
diff changeset
22 ;; Boston, MA 02110-1301, USA.
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
23
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
24 ;;; Commentary:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
25
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
26 ;; These functions provide a simple way to wash/clean html infected
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
27 ;; mails. Definitely do not work in all cases, but some improvement
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
28 ;; in readability is generally obtained. Formatting is only done in
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
29 ;; the buffer, so the next time you enter the article it will be
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
30 ;; "re-htmlized".
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
31 ;;
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
32 ;; The main function is `html2text'.
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
33
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
34 ;;; Code:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
35
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
36 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
37 ;; <Global variables>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
38 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
39
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
40 (eval-when-compile
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
41 (require 'cl))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
42
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
43 (defvar html2text-format-single-element-list '(("hr" . html2text-clean-hr)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
44
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
45 (defvar html2text-replace-list
57243
c5e16264557d Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-575
Miles Bader <miles@gnu.org>
parents: 56927
diff changeset
46 '(("&nbsp;" . " ") ("&gt;" . ">") ("&lt;" . "<") ("&quot;" . "\"")
c5e16264557d Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-575
Miles Bader <miles@gnu.org>
parents: 56927
diff changeset
47 ("&amp;" . "&") ("&apos;" . "'"))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
48 "The map of entity to text.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
49
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
50 This is an alist were each element is a dotted pair consisting of an
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
51 old string, and a replacement string. This replacement is done by the
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
52 function `html2text-substitute' which basically performs a
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
53 `replace-string' operation for every element in the list. This is
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
54 completely verbatim - without any use of REGEXP.")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
55
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
56 (defvar html2text-remove-tag-list
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
57 '("html" "body" "p" "img" "dir" "head" "div" "br" "font" "title" "meta")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
58 "A list of removable tags.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
59
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
60 This is a list of tags which should be removed, without any
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
61 formatting. Note that tags in the list are presented *without*
68932
dc25f1fd56f5 (html2text-remove-tag-list): Fix typo in docstring.
Juanma Barranquero <lekktu@gmail.com>
parents: 68633
diff changeset
62 any \"<\" or \">\". All occurrences of a tag appearing in this
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
63 list are removed, irrespective of whether it is a closing or
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
64 opening tag, or if the tag has additional attributes. The
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
65 deletion is done by the function `html2text-remove-tags'.
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
66
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
67 For instance the text:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
68
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
69 \"Here comes something <font size\"+3\" face=\"Helvetica\"> big </font>.\"
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
70
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
71 will be reduced to:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
72
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
73 \"Here comes something big.\"
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
74
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
75 If this list contains the element \"font\".")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
76
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
77 (defvar html2text-format-tag-list
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
78 '(("b" . html2text-clean-bold)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
79 ("strong" . html2text-clean-bold)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
80 ("u" . html2text-clean-underline)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
81 ("i" . html2text-clean-italic)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
82 ("em" . html2text-clean-italic)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
83 ("blockquote" . html2text-clean-blockquote)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
84 ("a" . html2text-clean-anchor)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
85 ("ul" . html2text-clean-ul)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
86 ("ol" . html2text-clean-ol)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
87 ("dl" . html2text-clean-dl)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
88 ("center" . html2text-clean-center))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
89 "An alist of tags and processing functions.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
90
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
91 This is an alist where each dotted pair consists of a tag, and then
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
92 the name of a function to be called when this tag is found. The
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
93 function is called with the arguments p1, p2, p3 and p4. These are
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
94 demontrated below:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
95
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
96 \"<b> This is bold text </b>\"
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
97 ^ ^ ^ ^
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
98 | | | |
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
99 p1 p2 p3 p4
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
100
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
101 Then the called function will typically format the text somewhat and
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
102 remove the tags.")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
103
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
104 (defvar html2text-remove-tag-list2 '("li" "dt" "dd" "meta")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
105 "Another list of removable tags.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
106
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
107 This is a list of tags which are removed similarly to the list
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
108 `html2text-remove-tag-list' - but these tags are retained for the
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
109 formatting, and then moved afterward.")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
110
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
111 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
112 ;; </Global variables>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
113 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
114
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
115 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
116 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
117
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
118 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
119 ;; <Utility functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
120 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
121
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
122
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
123 (defun html2text-replace-string (from-string to-string min max)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
124 "Replace FROM-STRING with TO-STRING in region from MIN to MAX."
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
125 (goto-char min)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
126 (let ((delta (- (string-width to-string) (string-width from-string)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
127 (change 0))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
128 (while (search-forward from-string max t)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
129 (replace-match to-string)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
130 (setq change (+ change delta)))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
131 change))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
132
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
133 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
134 ;; </Utility functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
135 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
136
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
137 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
138 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
139
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
140 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
141 ;; <Functions related to attributes> i.e. <font size=+3>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
142 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
143
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
144 (defun html2text-attr-value (list attribute)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
145 "Get value of ATTRIBUTE from LIST."
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
146 (nth 1 (assoc attribute list)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
147
57923
d7def5572cf3 Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-668
Miles Bader <miles@gnu.org>
parents: 57856
diff changeset
148 (defun html2text-get-attr (p1 p2)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
149 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
150 (re-search-forward " +[^ ]" p2 t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
151 (let* ((attr-string (buffer-substring-no-properties (1- (point)) (1- p2)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
152 (tmp-list (split-string attr-string))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
153 (attr-list)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
154 (counter 0)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
155 (prev (car tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
156 (this (nth 1 tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
157 (next (nth 2 tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
158 (index 1))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
159
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
160 (cond
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
161 ;; size=3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
162 ((string-match "[^ ]=[^ ]" prev)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
163 (let ((attr (nth 0 (split-string prev "=")))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
164 (value (nth 1 (split-string prev "="))))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
165 (setq attr-list (cons (list attr value) attr-list))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
166 ;; size= 3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
167 ((string-match "[^ ]=\\'" prev)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
168 (setq attr-list (cons (list (substring prev 0 -1) this) attr-list))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
169
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
170 (while (< index (length tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
171 (cond
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
172 ;; size=3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
173 ((string-match "[^ ]=[^ ]" this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
174 (let ((attr (nth 0 (split-string this "=")))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
175 (value (nth 1 (split-string this "="))))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
176 (setq attr-list (cons (list attr value) attr-list))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
177 ;; size =3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
178 ((string-match "\\`=[^ ]" this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
179 (setq attr-list (cons (list prev (substring this 1)) attr-list)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
180 ;; size= 3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
181 ((string-match "[^ ]=\\'" this)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
182 (setq attr-list (cons (list (substring this 0 -1) next) attr-list)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
183 ;; size = 3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
184 ((string= "=" this)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
185 (setq attr-list (cons (list prev next) attr-list))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
186 (setq index (1+ index))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
187 (setq prev this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
188 (setq this next)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
189 (setq next (nth (1+ index) tmp-list)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
190 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
191 ;; Tags with no accompanying "=" i.e. value=nil
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
192 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
193 (setq prev (car tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
194 (setq this (nth 1 tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
195 (setq next (nth 2 tmp-list))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
196 (setq index 1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
197
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
198 (when (and (not (string-match "=" prev))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
199 (not (string= (substring this 0 1) "=")))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
200 (setq attr-list (cons (list prev nil) attr-list)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
201 (while (< index (1- (length tmp-list)))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
202 (when (and (not (string-match "=" this))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
203 (not (or (string= (substring next 0 1) "=")
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
204 (string= (substring prev -1) "="))))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
205 (setq attr-list (cons (list this nil) attr-list)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
206 (setq index (1+ index))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
207 (setq prev this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
208 (setq this next)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
209 (setq next (nth (1+ index) tmp-list)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
210
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
211 (when (and this
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
212 (not (string-match "=" this))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
213 (not (string= (substring prev -1) "=")))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
214 (setq attr-list (cons (list this nil) attr-list)))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
215 ;; return - value
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
216 attr-list))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
217
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
218 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
219 ;; </Functions related to attributes>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
220 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
221
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
222 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
223 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
224
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
225 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
226 ;; <Functions to be called to format a tag-pair>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
227 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
228 (defun html2text-clean-list-items (p1 p2 list-type)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
229 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
230 (let ((item-nr 0)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
231 (items 0))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
232 (while (re-search-forward "<li>" p2 t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
233 (setq items (1+ items)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
234 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
235 (while (< item-nr items)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
236 (setq item-nr (1+ item-nr))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
237 (re-search-forward "<li>" (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
238 (cond
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
239 ((string= list-type "ul") (insert " o "))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
240 ((string= list-type "ol") (insert (format " %s: " item-nr)))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
241 (t (insert " x "))))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
242
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
243 (defun html2text-clean-dtdd (p1 p2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
244 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
245 (let ((items 0)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
246 (item-nr 0))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
247 (while (re-search-forward "<dt>" p2 t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
248 (setq items (1+ items)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
249 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
250 (while (< item-nr items)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
251 (setq item-nr (1+ item-nr))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
252 (re-search-forward "<dt>\\([ ]*\\)" (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
253 (when (match-string 1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
254 (delete-region (point) (- (point) (string-width (match-string 1)))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
255 (let ((def-p1 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
256 (def-p2 0))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
257 (re-search-forward "\\([ ]*\\)\\(</dt>\\|<dd>\\)" (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
258 (if (match-string 1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
259 (progn
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
260 (let* ((mw1 (string-width (match-string 1)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
261 (mw2 (string-width (match-string 2)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
262 (mw (+ mw1 mw2)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
263 (goto-char (- (point) mw))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
264 (delete-region (point) (+ (point) mw1))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
265 (setq def-p2 (point))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
266 (setq def-p2 (- (point) (string-width (match-string 2)))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
267 (put-text-property def-p1 def-p2 'face 'bold)))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
268
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
269 (defun html2text-delete-tags (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
270 (delete-region p1 p2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
271 (delete-region (- p3 (- p2 p1)) (- p4 (- p2 p1))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
272
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
273 (defun html2text-delete-single-tag (p1 p2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
274 (delete-region p1 p2))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
275
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
276 (defun html2text-clean-hr (p1 p2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
277 (html2text-delete-single-tag p1 p2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
278 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
279 (newline 1)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
280 (insert (make-string fill-column ?-)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
281
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
282 (defun html2text-clean-ul (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
283 (html2text-delete-tags p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
284 (html2text-clean-list-items p1 (- p3 (- p1 p2)) "ul"))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
285
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
286 (defun html2text-clean-ol (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
287 (html2text-delete-tags p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
288 (html2text-clean-list-items p1 (- p3 (- p1 p2)) "ol"))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
289
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
290 (defun html2text-clean-dl (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
291 (html2text-delete-tags p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
292 (html2text-clean-dtdd p1 (- p3 (- p1 p2))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
293
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
294 (defun html2text-clean-center (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
295 (html2text-delete-tags p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
296 (center-region p1 (- p3 (- p2 p1))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
297
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
298 (defun html2text-clean-bold (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
299 (put-text-property p2 p3 'face 'bold)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
300 (html2text-delete-tags p1 p2 p3 p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
301
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
302 (defun html2text-clean-title (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
303 (put-text-property p2 p3 'face 'bold)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
304 (html2text-delete-tags p1 p2 p3 p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
305
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
306 (defun html2text-clean-underline (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
307 (put-text-property p2 p3 'face 'underline)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
308 (html2text-delete-tags p1 p2 p3 p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
309
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
310 (defun html2text-clean-italic (p1 p2 p3 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
311 (put-text-property p2 p3 'face 'italic)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
312 (html2text-delete-tags p1 p2 p3 p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
313
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
314 (defun html2text-clean-font (p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
315 (html2text-delete-tags p1 p2 p3 p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
316
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
317 (defun html2text-clean-blockquote (p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
318 (html2text-delete-tags p1 p2 p3 p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
319
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
320 (defun html2text-clean-anchor (p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
321 ;; If someone can explain how to make the URL clickable I will surely
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
322 ;; improve upon this.
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
323 ;; Maybe `goto-addr.el' can be used here.
57923
d7def5572cf3 Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-668
Miles Bader <miles@gnu.org>
parents: 57856
diff changeset
324 (let* ((attr-list (html2text-get-attr p1 p2))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
325 (href (html2text-attr-value attr-list "href")))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
326 (delete-region p1 p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
327 (when href
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
328 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
329 (insert (substring href 1 -1 ))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
330 (put-text-property p1 (point) 'face 'bold))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
331
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
332 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
333 ;; </Functions to be called to format a tag-pair>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
334 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
335
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
336 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
337 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
338
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
339 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
340 ;; <Functions to be called to fix up paragraphs>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
341 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
342
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
343 (defun html2text-fix-paragraph (p1 p2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
344 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
345 (let ((has-br-line)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
346 (refill-start)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
347 (refill-stop))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
348 (when (re-search-forward "<br>$" p2 t)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
349 (goto-char p1)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
350 (when (re-search-forward ".+[^<][^b][^r][^>]$" p2 t)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
351 (beginning-of-line)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
352 (setq refill-start (point))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
353 (goto-char p2)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
354 (re-search-backward ".+[^<][^b][^r][^>]$" refill-start t)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
355 (next-line 1)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
356 (end-of-line)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
357 ;; refill-stop should ideally be adjusted to
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
358 ;; accomodate the "<br>" strings which are removed
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
359 ;; between refill-start and refill-stop. Can simply
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
360 ;; be returned from my-replace-string
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
361 (setq refill-stop (+ (point)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
362 (html2text-replace-string
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
363 "<br>" ""
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
364 refill-start (point))))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
365 ;; (message "Point = %s refill-stop = %s" (point) refill-stop)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
366 ;; (sleep-for 4)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
367 (fill-region refill-start refill-stop))))
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
368 (html2text-replace-string "<br>" "" p1 p2))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
369
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
370 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
371 ;; This one is interactive ...
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
372 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
373 (defun html2text-fix-paragraphs ()
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
374 "This _tries_ to fix up the paragraphs - this is done in quite a ad-hook
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
375 fashion, quite close to pure guess-work. It does work in some cases though."
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
376 (interactive)
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
377 (goto-char (point-min))
62943
11d53dd5abd9 Revision: miles@gnu.org--gnu-2005/emacs--cvs-trunk--0--patch-345
Miles Bader <miles@gnu.org>
parents: 57923
diff changeset
378 (while (re-search-forward "^<br>$" nil t)
11d53dd5abd9 Revision: miles@gnu.org--gnu-2005/emacs--cvs-trunk--0--patch-345
Miles Bader <miles@gnu.org>
parents: 57923
diff changeset
379 (delete-region (match-beginning 0) (match-end 0)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
380 ;; Removing lonely <br> on a single line, if they are left intact we
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
381 ;; dont have any paragraphs at all.
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
382 (goto-char (point-min))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
383 (while (not (eobp))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
384 (let ((p1 (point)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
385 (forward-paragraph 1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
386 ;;(message "Kaller fix med p1=%s p2=%s " p1 (1- (point))) (sleep-for 5)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
387 (html2text-fix-paragraph p1 (1- (point)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
388 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
389 (when (not (eobp))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
390 (forward-paragraph 1)))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
391
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
392 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
393 ;; </Functions to be called to fix up paragraphs>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
394 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
395
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
396 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
397 ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
398
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
399 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
400 ;; <Interactive functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
401 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
402
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
403 (defun html2text-remove-tags (tag-list)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
404 "Removes the tags listed in the list `html2text-remove-tag-list'.
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
405 See the documentation for that variable."
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
406 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
407 (dolist (tag tag-list)
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
408 (goto-char (point-min))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
409 (while (re-search-forward (format "\\(</?%s[^>]*>\\)" tag) (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
410 (delete-region (match-beginning 0) (match-end 0)))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
411
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
412 (defun html2text-format-tags ()
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
413 "See the variable `html2text-format-tag-list' for documentation."
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
414 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
415 (dolist (tag-and-function html2text-format-tag-list)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
416 (let ((tag (car tag-and-function))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
417 (function (cdr tag-and-function)))
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
418 (goto-char (point-min))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
419 (while (re-search-forward (format "\\(<%s\\( [^>]*\\)?>\\)" tag)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
420 (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
421 (let ((p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
422 (p2 (point))
57923
d7def5572cf3 Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-668
Miles Bader <miles@gnu.org>
parents: 57856
diff changeset
423 (p3) (p4))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
424 (search-backward "<" (point-min) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
425 (setq p1 (point))
73627
e7c14a8e6688 Merge from gnus--rel--5.10
Miles Bader <miles@gnu.org>
parents: 68932
diff changeset
426 (unless (search-forward (format "</%s>" tag) (point-max) t)
e7c14a8e6688 Merge from gnus--rel--5.10
Miles Bader <miles@gnu.org>
parents: 68932
diff changeset
427 (goto-char p2)
e7c14a8e6688 Merge from gnus--rel--5.10
Miles Bader <miles@gnu.org>
parents: 68932
diff changeset
428 (insert (format "</%s>" tag)))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
429 (setq p4 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
430 (search-backward "</" (point-min) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
431 (setq p3 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
432 (funcall function p1 p2 p3 p4)
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
433 (goto-char p1))))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
434
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
435 (defun html2text-substitute ()
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
436 "See the variable `html2text-replace-list' for documentation."
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
437 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
438 (dolist (e html2text-replace-list)
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
439 (goto-char (point-min))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
440 (let ((old-string (car e))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
441 (new-string (cdr e)))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
442 (html2text-replace-string old-string new-string (point-min) (point-max)))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
443
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
444 (defun html2text-format-single-elements ()
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
445 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
446 (dolist (tag-and-function html2text-format-single-element-list)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
447 (let ((tag (car tag-and-function))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
448 (function (cdr tag-and-function)))
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
449 (goto-char (point-min))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
450 (while (re-search-forward (format "\\(<%s\\( [^>]*\\)?>\\)" tag)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
451 (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
452 (let ((p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
453 (p2 (point)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
454 (search-backward "<" (point-min) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
455 (setq p1 (point))
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
456 (funcall function p1 p2))))))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
457
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
458 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
459 ;; Main function
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
460 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
461
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
462 ;;;###autoload
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
463 (defun html2text ()
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
464 "Convert HTML to plain text in the current buffer."
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
465 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
466 (save-excursion
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
467 (let ((case-fold-search t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
468 (buffer-read-only))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
469 (html2text-remove-tags html2text-remove-tag-list)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
470 (html2text-format-tags)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
471 (html2text-remove-tags html2text-remove-tag-list2)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
472 (html2text-substitute)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
473 (html2text-format-single-elements)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
474 (html2text-fix-paragraphs))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
475
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
476 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
477 ;; </Interactive functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
478 ;;
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
479 (provide 'html2text)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
480 ;;; arch-tag: e9e57b79-35d4-4de1-a647-e7e01fe56d1e
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
481 ;;; html2text.el ends here