diff options
Diffstat (limited to 'abcnews.go.com.txt')
| -rw-r--r-- | abcnews.go.com.txt | 27 |
1 files changed, 27 insertions, 0 deletions
diff --git a/abcnews.go.com.txt b/abcnews.go.com.txt new file mode 100644 index 0000000..8d36735 --- /dev/null +++ b/abcnews.go.com.txt @@ -0,0 +1,27 @@ +title: //h1[@class='headline'] +body: //div[@id='storyText'] +# for video entries +body: //img[@id='ff-img'] | //div[@id='meta']//div[contains(@class, 'overview')] +author: //div[@class='byline'] +date: //div[@class='date'] +strip: //*[@id='date_partner'] + +strip: //div[@class='breadcrumb'] +strip: //div[contains(@class,'show_tools')] +strip: //div[@id='sponsoredByAd'] +strip: //div[contains(@class,'rel_container')] +strip: //p[a[starts-with(@href, 'http://www.twitter.com')]] +strip: //p[a[starts-with(@href, 'http://www.facebook.com')]] +strip: //p[contains(., 'Click here to return to')] +#strip_id_or_class: media +strip_id_or_class: mediaplayer + +replace_string(<link rel="image_src" href="http): <img id="ff-img" src="http + +prune: no + +single_page_link: concat(//li[@class='pager']//a/@href, '&singlePage=true') + +test_url: http://abcnews.go.com/Politics/newt-gingrich-rocky-rollout-presidential-campaign-recover/story?id=13632744 +# multi-page +test_url: http://abcnews.go.com/Blotter/family-freed-american-hostage-somalia-seals-obama/story?id=15439544
\ No newline at end of file |