scraper.rb

require 'kinto_box'
require 'json'
require 'nokogiri'
require 'httparty'

# Load kinto connection information from datastore-config.json
config = JSON.parse(File.read('datastore-config.json'))

# Create a new Kinto connection
kinto = KintoBox.new(config['kinto_server'])

# Get the bucket object for this project
BUCKET = kinto.bucket(config['bucket'])

# Model of the record to be stored
class Movie
  @@collection = BUCKET.collection('movies')
  def initialize(html, rank)
    @rank = rank
    @title = html.css('.lister-item-content').css('h3').css('a')[0].text
    @rating = html.css('.ratings-imdb-rating').css('strong').text
    @poster_img_url = html.css('.lister-item-image').css('a').css('img')[0]['src']
  end

  def object_hash
    {
      'rank' => @rank,
      'title' => @title,
      'rating' => @rating,
      'posterImgUrl' => @poster_img_url
    }
  end

  def push_to_kinto
    @@collection.create_record(object_hash)
  end

  def self.delete_all_in_kinto
    @@collection.delete_records
  end
end

# Scraper ruby script
puts 'scraping...'
url = 'http://www.imdb.com/search/title?groups=top_100&sort=user_rating'
page_html = HTTParty.get(url)
parsed_html = Nokogiri::HTML(page_html)


movies = []
Movie.delete_all_in_kinto
parsed_html.css('.lister-list .lister-item').each_with_index do |movie, i|
  new_movie = Movie.new(movie, i + 1)
  movies.push(new_movie)
  new_movie.push_to_kinto
end