#!/bin/sh
#
# This script removes the HTML formatting from a file. If the file was designed
# with such use in mind and was properly formatted besides HTML (such as the README
# file for ttf2pt1) it will look good as a plain text file.
#
# This script supports a very limited set of HTML formatting. Everything that
# goes before <BODY> is removed.  Any lines that
# contain only the HTML formatting or start with "<!" or contain only ">"
# are completely removed. Then all the in-line formatting is removed.
# Then "&nbsp;", "&lt;", "&gt;" are changed to " ", "<", ">".

sed '1,/<BODY>/d;/^<!/d;s/^</< </;s/> *$/>>/;s/<[^<>]*>//g;
/^< *>$/d;/^>>$/d;s/^< //;s/>$//;
s/&nbsp;/ /g;s/&lt;/</g;s/&gt;/>/g;'
