inc/3rdparty/site_config/standard/bostonglobe.com.txt

   1 # NOTE:  If testing this configuration yields bad results, including junk text like "Try BostonGlobe.com today" and "THIS STORY APPEARED IN", please replace the Test URL with a current-day headline link from bostonglobe.com.
   2
   3 title: //div[@class="header"]/h1
   4 author: substring-after(//div[@class="byline"]/h2[@class="author"],"By ")
   5 date: //div[@class="byline"]/p[last()]
   6 body: //div[@class="article-body"]
   7
   8 strip_id_or_class: aside
   9 strip_id_or_class: promo
  10 strip_id_or_class: skip-nav
  11 strip_id_or_class: article-more
  12 strip_id_or_class: article-bar
  13
  14 # This removes image captions.  If the parser starts saving images from bostonglobe.com (currently, it does not), then this directive should be removed.
  15 strip_id_or_class: figure
  16 test_url: http://bostonglobe.com/news/nation/2012/03/17/illinois-primary-could-pivotal/PsDzFZqvhEYyXbOcF9FOkO/story.html