loudatui.rb

require 'net/http'
require 'nokogiri'
###
# Lsong
# [email protected]
# http://lsong.org
# 
# MIT LICENSE
# http://lsong.mit-license.org/
# 
# RUNTIME
# ==============
# $ ruby -v
# ruby 2.0.0p247 (2013-06-27 revision 41674) [x86_64-darwin12.4.0]
#


def fetch_page(num)
  puts "Now index from page #{num}"
  base_url =  "http://loudatui.com"
  #images in page
  images_url = []
  #make a uri
  uri = URI.parse(base_url)
  Net::HTTP.start(uri.host,uri.port) do |http|
    resp = http.get("/?page=#{num}")
    #parse xml
    doc = Nokogiri::XML(resp.body)
    #find all <img class="img" />
    images =  doc.css("img.img")
    images.each do |img|
      #get image src attribute
      images_url << img.attr('src')
    end
  end
  images_url
end


def fetch_site
  #
  puts "Downloading index , will take a minute\n" 
  puts "#{'-' * 50}\n"

  n = 1 #page begin from 1
  #temp var
  images_url = []
  #fetch all page
  while (images_url = fetch_page(n)).count > 0 do
    #merge arrays
    images_url.each do |img_url|
	download_image img_url
    end
    n = n + 1
  end
end



def download_image(url)  
  #get a big pic , not small .
  url = url.gsub('/small','/large')
  #file name
  fname = /[0-9a-z]{22,32}/.match(url)[0] || Time.now.to_i.to_s
  # .ext
  fname = "#{fname}.jpg"
  #break , if file already exist . 
  return if File.exist? fname
  File.open(fname,'wb') do |f|
    uri = URI.parse(url)
    Net::HTTP.start(uri.host,uri.port) do |http|
      resp =  http.get(uri.path)
      f.write resp.body
      puts "download image file : #{fname}"
    end
  end
end

def main
  puts "Image downloader for loudatui.com\n"
  puts "#{'=' * 50}\n"
  fetch_site
  #have a fun .
  puts "finished"
end


main