CGI.escapeHTML is pretty bad, but CGI.unescapeHTML is completely borked. For example:
CGI.escapeHTML
CGI.unescapeHTML
require \'cgi\' CGI.unescapeHTML(\'…\
require 'rubygems' require 'hpricot' Hpricot('…', :xhtml_strict => true).to_plain_text
Though you might have to fiddle around with the character encoding.