BBC headline scraper. Requires Hpricot.

bbc.rb

require 'rubygems'
require 'hpricot'
require 'open-uri'

BBC_URL = 'http://news.bbc.co.uk'
SELECTORS = '.ticker_content_anchor , #other-top-stories .story, #third-story .story, #most-popular .story, #featured-site-top-stories---democracy-live .story, #more-from-bbc-news .story, #featured-site-top-stories---bbc-sport .story, #geo-uk-news-digest li .story, #also-in-the-news .story, #second-story .story, #top-story .story'

bbc = open(BBC_URL)

links = Hpricot(bbc).search(SELECTORS)

links.each do |link|
  title = link.inner_text.strip
  url = link['href']
  unless title.empty? or url.empty?
    puts title + ": " + (url.index('http://') == 0 ? url : BBC_URL + url)
  end
end