Scrapping with nokogiri

gistfile1.rb

require 'net/http'
require "nokogiri"
require 'open-uri'

module Scrapper
  def get_images_from(url)
    images = []
    doc = Nokogiri::HTML(open(url))
    doc.xpath("/html/body//img[@src[ not(contains(.,'\.gif'))
                                    and contains(.,'://')
                                    and not(contains(.,'ads.')
                                    or contains(.,'ad.')
                                    or contains(.,'?'))]][1]").each do |img|
      image = img.attributes["src"].value
      size = image_size_from(image)
      images << image
    end
    images
  end

  def image_size_from(asset)
    url = URI.parse asset
    res = Net::HTTP.start(url.host, url.port) do |http|
      response = http.request_head(asset)
      return response['content-length'].to_i
    end
  end
end