diff options
Diffstat (limited to 'businessinsider.com.txt')
| -rw-r--r-- | businessinsider.com.txt | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/businessinsider.com.txt b/businessinsider.com.txt index f847687..39eb742 100644 --- a/businessinsider.com.txt +++ b/businessinsider.com.txt @@ -1,12 +1,16 @@ title://div[@class="sl-layout-post"]/h1 -body: //div[contains(@class, 'post-content') or contains(@class, 'KonaBody')] +body: //div[contains(@class, 'post-content') or contains(@class, 'slide-module') or contains(@class, 'KonaBody')] strip: //div[contains(@class, "post-sidebar")] strip: //div[@id='related-links'] +strip: //div[@class='related-links-container'] +strip: //p[@class='source'] author://div[@class="byline"]/a date://div[@class="byline"]/span[@class="date"] prune: no +single_page_link: //a[contains(text(), 'View as one page')] + strip://*[contains(@class,'sponsored-text')] strip: //div[@id='post_footer'] -test_url: http://www.businessinsider.com/microsoft-just-put-one-of-its-hardcore-technical-geniuses-on-xbox-2012-1
\ No newline at end of file +test_url: http://www.businessinsider.com/microsoft-just-put-one-of-its-hardcore-technical-geniuses-on-xbox-2012-1 |