body: //div[@id="main"] title: //div[@id="main"]/h3 # Remove ‘Review’ and ‘Wii’. strip: //div[@class="badge"] # Remove duplicate title and country flag. strip: //h3 # Commented out below are attempts to extract the author and date, which did not work. # author: //p[@class="extra "]/a # date: //p[@class="extra "]/span[@class="when"] test_url: http://www.nintendoworldreport.com/review/28400