1 title: //meta[@name='og:title']/@content
2 date: //meta[@name='created']/@content
3 body: //div[@class="StoryBody" or @class="storyTeaser"]
5 replace_string(<p></p>): <br /><br />
7 test_url: http://www.thebostonchannel.com/slideshow/news/28210648/detail.html