annotate lisp/gnus/html2text.el @ 57856:df80d19d7a2e

Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660 Merge from gnus--rel--5.10 Patches applied: * miles@gnu.org--gnu-2004/gnus--rel--5.10--patch-65 Update from CVS 2004-11-01 Reiner Steib <Reiner.Steib@gmx.de> * lisp/gnus/gnus-msg.el (gnus-summary-resend-default-address): Add :version. * lisp/gnus/tls.el (tls-process-connection-type, tls-success) (tls-certtool-program): Add :version. * lisp/gnus/starttls.el (starttls-gnutls-program, starttls-use-gnutls) (starttls-extra-arguments, starttls-process-connection-type) (starttls-connect, starttls-failure, starttls-success): * lisp/gnus/spam-stat.el (spam-stat): Add :version. * lisp/gnus/sieve.el (sieve): Add :version. * lisp/gnus/sha1.el (sha1): Added :version. (sha1-use-external): Removed redundant version. * lisp/gnus/nnmail.el (nnmail-split-fancy-with-parent-ignore-groups) (nnmail-cache-ignore-groups, nnmail-spool-hook) (nnmail-split-fancy-match-partial-words) (nnmail-split-lowercase-expanded): * lisp/gnus/nndiary.el (nndiary): Add :version. * lisp/gnus/mml2015.el (mml2015-unabbrev-trust-alist): Add :version. * lisp/gnus/mml-sec.el (mml-default-sign-method) (mml-default-encrypt-method, mml-signencrypt-style-alist): Add :version. * lisp/gnus/mm-uu.el (mm-uu-diff-groups-regexp): Add :version. * lisp/gnus/mm-url.el (mm-url-use-external, mm-url-program) (mm-url-arguments): Add :version. * lisp/gnus/mm-decode.el (mm-inline-text-html-with-w3m-keymap) (mm-attachment-file-modes, mm-decrypt-option) (mm-w3m-safe-url-regexp): Add :version. * lisp/gnus/message.el (message-cite-prefix-regexp) (message-sendmail-envelope-from, message-minibuffer-local-map) (message-user-fqdn, message-completion-alist): Add :version. * lisp/gnus/gnus-win.el (gnus-configure-windows-hook) (gnus-use-frames-on-any-display): Add :version. * lisp/gnus/gnus-art.el (gnus-article-address-banner-alist) (gnus-treat-unsplit-urls, gnus-treat-unfold-headers) (gnus-treat-from-picon, gnus-treat-mail-picon) (gnus-treat-x-pgp-sig): Add :version. * lisp/gnus/gnus-sum.el (gnus-spam-mark, gnus-recent-mark) (gnus-undownloaded-mark, gnus-summary-article-move-hook) (gnus-summary-article-delete-hook) (gnus-summary-display-while-building): Add :version. * lisp/gnus/gnus-start.el (gnus-subscribe-newsgroup-hooks) (gnus-get-top-new-news-hook):Add :version. * lisp/gnus/gnus-srvr.el (gnus-server-agent-face, gnus-server-opened-face) (gnus-server-closed-face, gnus-server-denied-face): Add :version. * lisp/gnus/gnus-registry.el (gnus-registry): Add :version. * lisp/gnus/gnus-spec.el (gnus-use-correct-string-widths) (gnus-make-format-preserve-properties): Add :version. * lisp/gnus/gnus.el (gnus-group-charter-alist) (gnus-group-fetch-control-use-browse-url) (gnus-install-group-spam-parameters): Add :version. * lisp/gnus/gnus-diary.el (gnus-diary): Add :version. * lisp/gnus/gnus-delay.el (gnus-delay): Add :version. * lisp/gnus/gnus-cite.el (gnus-cite-unsightly-citation-regexp) (gnus-cite-ignore-quoted-from, gnus-cite-attribution-face) (gnus-cite-blank-line-after-header, gnus-article-boring-faces): Add :version. * lisp/gnus/gnus-agent.el (gnus-agent-max-fetch-size) (gnus-agent-enable-expiration, gnus-agent-queue-mail) (gnus-agent-prompt-send-queue): Add :version. * lisp/gnus/deuglify.el (gnus-outlook-deuglify): Add :version. * lisp/gnus/html2text.el: Beautify code. Improve doc strings. Some checkdoc cleanup. (html2text-get-attr, html2text-fix-paragraph): Simplify code. (html2text-format-tag-list): Added "strong" and "em". From "Alfred M. Szmidt" <ams@kemisten.nu> (tiny change).
author Miles Bader <miles@gnu.org>
date Mon, 01 Nov 2004 23:06:36 +0000
parents 6a65cb24e1be
children d7def5572cf3
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
57243
c5e16264557d Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-575
Miles Bader <miles@gnu.org>
parents: 56927
diff changeset
2 ;; Copyright (C) 2002, 2003, 2004 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
3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
4 ;; 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
5
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
6 ;; 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
7
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
8 ;; 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
9 ;; 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
10 ;; 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
11 ;; any later version.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
12
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
13 ;; 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
14 ;; 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
15 ;; 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
16 ;; 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
17
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
18 ;; 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
19 ;; along with GNU Emacs; see the file COPYING. If not, write to the
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
20 ;; Free Software Foundation, Inc., 59 Temple Place - Suite 330,
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
21 ;; Boston, MA 02111-1307, USA.
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
22
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
23 ;;; Commentary:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
24
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
25 ;; 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
26 ;; 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
27 ;; 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
28 ;; 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
29 ;; "re-htmlized".
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
30 ;;
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
31 ;; 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
32
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
33 ;;; Code:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
34
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 ;; <Global variables>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
37 ;;
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 (eval-when-compile
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
40 (require 'cl))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
41
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
42 (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
43
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
44 (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
45 '(("&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
46 ("&amp;" . "&") ("&apos;" . "'"))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
47 "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
48
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
49 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
50 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
51 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
52 `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
53 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
54
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
55 (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
56 '("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
57 "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
58
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
59 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
60 formatting. Note that tags in the list are presented *without*
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
61 any \"<\" or \">\". All occurences of a tag appearing in this
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
62 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
63 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
64 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
65
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
66 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
67
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
68 \"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
69
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
70 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
71
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
72 \"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
73
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
74 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
75
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
76 (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
77 '(("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
78 ("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
79 ("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
80 ("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
81 ("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
82 ("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
83 ("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
84 ("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
85 ("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
86 ("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
87 ("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
88 "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
89
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
90 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
91 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
92 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
93 demontrated below:
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
94
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
95 \"<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
96 ^ ^ ^ ^
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 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
99
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
100 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
101 remove the tags.")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
102
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
103 (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
104 "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
105
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
106 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
107 `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
108 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
109
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 ;; </Global variables>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
112 ;;
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 ;; <Utility functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
119 ;;
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
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
122 (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
123 "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
124 (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
125 (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
126 (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
127 (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
128 (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
129 (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
130 change))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
131
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 ;; </Utility functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
134 ;;
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 ;; <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
141 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
142
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
143 (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
144 "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
145 (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
146
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
147 (defun html2text-get-attr (p1 p2 tag)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
148 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
149 (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
150 (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
151 (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
152 (attr-list)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
153 (counter 0)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
154 (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
155 (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
156 (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
157 (index 1))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
158
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
159 (cond
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
160 ;; size=3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
161 ((string-match "[^ ]=[^ ]" prev)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
162 (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
163 (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
164 (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
165 ;; size= 3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
166 ((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
167 (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
168
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
169 (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
170 (cond
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
171 ;; size=3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
172 ((string-match "[^ ]=[^ ]" this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
173 (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
174 (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
175 (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
176 ;; size =3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
177 ((string-match "\\`=[^ ]" this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
178 (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
179 ;; size= 3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
180 ((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
181 (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
182 ;; size = 3
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
183 ((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
184 (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
185 (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
186 (setq prev this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
187 (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
188 (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
189 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
190 ;; 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
191 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
192 (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
193 (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
194 (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
195 (setq index 1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
196
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
197 (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
198 (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
199 (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
200 (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
201 (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
202 (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
203 (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
204 (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
205 (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
206 (setq prev this)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
207 (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
208 (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
209
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
210 (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
211 (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
212 (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
213 (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
214 ;; return - value
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
215 attr-list))
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
216
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 ;; </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
219 ;;
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 ;; <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
226 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
227 (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
228 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
229 (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
230 (items 0))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
231 (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
232 (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
233 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
234 (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
235 (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
236 (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
237 (cond
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
238 ((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
239 ((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
240 (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
241
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
242 (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
243 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
244 (let ((items 0)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
245 (item-nr 0))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
246 (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
247 (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
248 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
249 (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
250 (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
251 (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
252 (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
253 (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
254 (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
255 (def-p2 0))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
256 (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
257 (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
258 (progn
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
259 (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
260 (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
261 (mw (+ mw1 mw2)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
262 (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
263 (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
264 (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
265 (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
266 (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
267
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
268 (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
269 (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
270 (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
271
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
272 (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
273 (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
274
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
275 (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
276 (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
277 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
278 (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
279 (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
280
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
281 (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
282 (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
283 (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
284
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
285 (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
286 (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
287 (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
288
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
289 (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
290 (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
291 (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
292
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
293 (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
294 (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
295 (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
296
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
297 (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
298 (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
299 (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
300
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
301 (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
302 (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
303 (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
304
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
305 (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
306 (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
307 (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
308
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
309 (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
310 (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
311 (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
312
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
313 (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
314 (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
315
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
316 (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
317 (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
318
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
319 (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
320 ;; 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
321 ;; 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
322 ;; Maybe `goto-addr.el' can be used here.
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
323 (let* ((attr-list (html2text-get-attr p1 p2 "a"))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
324 (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
325 (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
326 (when href
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
327 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
328 (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
329 (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
330
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 ;; </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
333 ;;
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 ;; <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
340 ;;
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 (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
343 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
344 (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
345 (refill-start)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
346 (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
347 (when (re-search-forward "<br>$" p2 t)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
348 (goto-char p1)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
349 (when (re-search-forward ".+[^<][^b][^r][^>]$" p2 t)
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
350 (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
351 (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
352 (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
353 (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
354 (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
355 (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
356 ;; 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
357 ;; 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
358 ;; 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
359 ;; 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
360 (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
361 (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
362 "<br>" ""
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
363 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
364 ;; (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
365 ;; (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
366 (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
367 (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
368
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 ;; 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
371 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
372 (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
373 "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
374 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
375 (interactive)
57723
6a65cb24e1be Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-640
Miles Bader <miles@gnu.org>
parents: 57243
diff changeset
376 (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
377 (replace-regexp "^<br>$" "")
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
378 ;; 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
379 ;; 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
380 (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
381 (while (not (eobp))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
382 (let ((p1 (point)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
383 (forward-paragraph 1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
384 ;;(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
385 (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
386 (goto-char p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
387 (when (not (eobp))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
388 (forward-paragraph 1)))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
389
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
390 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
391 ;; </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
392 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
393
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 ;; <Interactive functions>
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
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
401 (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
402 "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
403 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
404 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
405 (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
406 (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
407 (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
408 (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
409
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
410 (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
411 "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
412 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
413 (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
414 (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
415 (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
416 (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
417 (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
418 (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
419 (let ((p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
420 (p2 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
421 (p3) (p4)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
422 (attr (match-string 1)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
423 (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
424 (setq p1 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
425 (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
426 (setq p4 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
427 (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
428 (setq p3 (point))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
429 (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
430 (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
431
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
432 (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
433 "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
434 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
435 (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
436 (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
437 (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
438 (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
439 (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
440
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
441 (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
442 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
443 (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
444 (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
445 (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
446 (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
447 (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
448 (point-max) t)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
449 (let ((p1)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
450 (p2 (point)))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
451 (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
452 (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
453 (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
454
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
455 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
456 ;; Main function
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 ;;;###autoload
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
460 (defun html2text ()
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
461 "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
462 (interactive)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
463 (save-excursion
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
464 (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
465 (buffer-read-only))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
466 (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
467 (html2text-format-tags)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
468 (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
469 (html2text-substitute)
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
470 (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
471 (html2text-fix-paragraphs))))
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
472
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
473 ;;
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
474 ;; </Interactive functions>
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
475 ;;
57856
df80d19d7a2e Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-660
Miles Bader <miles@gnu.org>
parents: 57723
diff changeset
476 (provide 'html2text)
56927
55fd4f77387a Revision: miles@gnu.org--gnu-2004/emacs--cvs-trunk--0--patch-523
Miles Bader <miles@gnu.org>
parents:
diff changeset
477 ;;; 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
478 ;;; html2text.el ends here