Small Ruby script that demonstrates how to use Mechanize to scrape some product details from an array of product URLs from Zappos.com

zappos_product_scraper.rb

# http://nokogiri.org/Nokogiri/XML/Node.html#method-i-css

require 'mechanize'
require 'csv'

puts "Product Scraper!!!"
puts ' '

urls = [
  "http://www.zappos.com/seavees-teva-universal-sandal-concrete",
  "http://www.zappos.com/teva-bomber-sandal-dark-olive",
  "http://www.zappos.com/teva-jetter-cigar"]


file = "product_data.csv"
header = "title,sku,image,alt_images"
File.open(file, "w") do |csv|
  csv << header
  csv << "\n"


  (0..urls.length - 1).each do |index|
    puts urls[index]
    agent = Mechanize.new
    page = agent.get(urls[index])

    title = page.title
    title = title[0..title.index(' - ')].rstrip

    sku = page.search("#sku").inner_text
    sku = sku[4..sku.length-1]

    prod_image = page.search("#detailImage img").first
    alt_images = page.search("#productImages ul li a img")

    brand_text = page.search("#brandText").inner_text

    alt_images = alt_images.map { |x| x[:src] }.join("|")

    csv << [title, sku, prod_image[:src], "#{alt_images}"]
    csv << "\n"
  end

2.times { |x| puts "" }
puts "Done!"

end