prune: no body: //div[@class='post-body'] author: //p[@class='byline']//a date: substring-after(//div[@class='about']/p[2], 'Posted') strip: //div[@class='body']/div[@class='meta'] test_url: http://www.autoblog.com/2012/01/17/next-gen-bmw-x5-caught-again/