Download all images for a given web page

download_images.rb

require 'nokogiri'
require 'open-uri'
require 'uri'
require 'pathname'

pages = ['http://images.google.com/search?tbm=isch&hl=en&source=hp&q=dogs',
         'http://images.google.com/search?tbm=isch&hl=en&source=hp&q=cats',
         'http://images.google.com/search?tbm=isch&hl=en&source=hp&q=hats']

pages.each do |page|
  doc   = Nokogiri::HTML(open(page))
  links = doc.css('a')
  hrefs = links.map {|link| link.attribute('href').to_s}.uniq.sort.delete_if {|href| href.empty? || href =~ /PROOF/ || href =~ /THUMB/ ||  href !~ /(.jpg|.png)/i}
  
  p = URI(page)  
  
  base_url = "#{(p.scheme.nil? ? "http" : p.scheme )}://#{p.host}#{p.port.nil? ? "" : ":#{p.port}"}"
  
  hrefs.each do |h|
    l           = URI(h)
    base_name   = Pathname.new(l.path).basename
    remote_file = "#{base_url}#{p.path}#{l.path}"
    out_file    = "wedding_images/#{l.path}"
    
    puts "  Downloading #{remote_file}, to #{out_file}"
        
    open(out_file, 'wb') do |file|
      file << open(remote_file).read
    end
    
  end
  
  
end