scraper.rb

# Example: List of all sports of the Olympic Games
# HTML Structure:
#
# <div id="content">
#   <ul>
#     <li>
#       <a href="...">SPORT'S NAME</a>
#     </li>
#   </ul>  
# </div>

require 'nokogiri'
require 'restclient'
require 'json'

sports = {}
BASE_URL = 'http://www.olympic.org'
sports_list_url = "#{BASE_URL}/sports"
page = Nokogiri::HTML(RestClient.get(sports_list_url))

links = page.css('#content ul li a').select # Select all sport contained in <li>

# To format output into JSON
links.each do |link|
  sports[link.text.to_sym] = "#{BASE_URL}/#{link['href']}" 
end
sports.to_json