topnews.rb

# Scrape Google News
#
# Grab the top stories from google news and print them
require 'rubygems'
require 'hpricot'
require 'open-uri'

class GoogleNews
  def initialize
    @url = "http://news.google.com"
    @hp  = Hpricot(open(@url))
  end

  def top_stories
    # Look for potential top story links, sources and descriptions
    titles = (@hp/"div[@id='top-stories']//h2.title/a").collect {|title|
      title.inner_text.empty? ? nil : title.inner_text.strip
    }
    titles.compact!

    srcs   = (@hp/"div[@id='top-stories']//div[@class='sub-title']/span.source").collect {|src|
      src.inner_text.strip
    }

    descs  = (@hp/"div[@id='top-stories']//div.body/div.snippet").collect {|desc|
      desc.inner_text.strip
    }

    # Next filter out everything but the story title
    result = []
    i = 0;

    titles.each {|title|
      result[i] = {}
      result[i]['title'] = title
      i += 1
    }

    i = 0
    srcs.each { |src|
      result[i]['source'] = src
      i += 1
    }
    
    i = 0
    descs.each {|desc|
      result[i]['desc'] = desc
      i += 1
    }

    result
  end
end

i = 1
gnews = GoogleNews.new
gnews.top_stories.each do |story|
  puts "**** #{i}: ****\n#{story['title']}\n"
  print "Source: #{story['source']}\n\n"
  print "#{story['desc']}\n\n"
  i += 1
end