Scrape an XHTML document using Ruby (See related posts)

#!/usr/bin/ruby
# file: thegang.rb

require 'rexml/document'
include REXML

class TheGang
  def initialize()
  end
  
  def rssify()
    file = File.new('thegang.xml','r')
    doc = Document.new(file)
    rss_doc = Document.new
    root = Element.new('rss')
    rss_doc.add_element(root)
    
    doc.root.elements.each("body/div/ul/li/h2/a") do |node|    
      o_rssitem = Element.new('item')
      o_li = node.parent.parent
      
      o_rsstitle = Element.new('title')
      o_rsstitle.text = node.text.gsub(/[\n,' ']/,'')
      o_rssitem.add_element(o_rsstitle)
      
      o_rsshref_audio = Element.new('href_audio')
      o_rsshref_audio.text = node.attributes.get_attribute('href').to_s.gsub('amp;&','')      
      o_rssitem.add_element(o_rsshref_audio)
      
      o_rsshref = Element.new('href')
      o_rsshref.text = o_rsshref_audio.text.gsub('&from=audio','')      
      o_rssitem.add_element(o_rsshref)
      
      o_rssdate = Element.new('date')
      o_rssdate.text = "#{o_li.elements["p/span[1]"].text} #{o_li.elements["p/span[2]"].text}"
      o_rssitem.add_element(o_rssdate)
      rss_doc.root.add_element(o_rssitem)
      
    end

    file = File.new('thegang_rss.xml','w')
    file.puts rss_doc
    file.close
  end
end


if __FILE__ == $0
  gang = TheGang.new
  gang.rssify
end

<rss>
  <item><title>TheGangXII-II</title><href_audio>/gangitem/id=6501&amp;from=audio</href_audio><href>/gangitem/id=6501</href><date>Jan 25</date></item>
  <item><title>TheGangXII-I</title><href_audio>/gangitem/id=6499&amp;from=audio</href_audio><href>/gangitem/id=6499</href><date>Jan 25</date></item>
  <item><title>NewsGangLive01.24.08</title><href_audio>/gangitem/id=6445&amp;from=audio</href_audio><href>/gangitem/id=6445</href><date>Jan 24</date></item>
  <item><title>NewsGangLiveII</title><href_audio>/gangitem/id=6377&amp;from=audio</href_audio><href>/gangitem/id=6377</href><date>Jan 23</date></item>
  ...
</rss>