2 date: //meta[@name="date"]/@content
3 author: //div[@class="author-name" or @class="article-byline"]/a[1]
5 body: //section[@class="page"]
7 # remove 'From the Lab' and 'Recent posts' text
8 strip: //div[@class='blogLabel']
10 # remove byline and meta info
11 strip: //div[@class="article-meta"]
12 strip: //div[@class="author-info"]
14 #strip tags and categories
15 strip: //div[@class="department"]
17 #strip product cap links
18 strip: //div[@class="cap-main"]
19 strip: //div[@id="compare-lede"]
23 # copes less well with Review pages, seems fine for News
24 test_url: http://www.macworld.com/article/163184/2011/10/the_ipod_as_an_iconic_cultural_force.html