Scraping Trulia to find a house

house_hunt.rb

require 'mechanize'
require 'awesome_print'

BASE = "http://www.trulia.com"

PREFIX = "/for_sale/"

PRICE = "215000-262000_price/"

NL = "7_nl/"

CITY_LIST = ["Apex,NC;Cary,NC/", "28846,28847_nh/27612,27613,27617_zip/"]


House = Struct.new(:address, :price, :link, :detail)
@agent = Mechanize.new

def load_page(city)
  url = BASE + PREFIX + city + PRICE + NL
  ap url
  @agent.get(url) 
end

def find_houses(page)
  address_sections = page.search("div.property-data-elem")
  ap address_sections.size
  
  houses = []
  
  address_sections.each do |sectnode|
    address = sectnode.at_css("div.address_section").at_css("a").text.strip
    link = BASE + sectnode.at_css("div.address_section").at_css("a")['href']
    price = sectnode.at_css("div.price_section").text.strip
    house = House.new(address, price, link, "" )
    houses << house
  end

  houses 
end


####################################################

CITY_LIST.each do |city|

  page = load_page(city)

  houses = find_houses(page)
 
  houses.each do |house|
    puts house.address + "  " + house.price
    puts house.link + "\n\n"
  end

end