gmusic spider

gistfile1.rb

#!/usr/bin/env ruby
require 'rubygems'
require 'nokogiri'
require 'mechanize'
require 'cgi'

a = Mechanize.new { |agent|
  agent.user_agent_alias = 'Mac Safari'
}

# puts "What is your name?"
# $name = STDIN.gets
# puts "Hi "+$name

def log(str)
  system("echo #{str} >> downloads.log")
end
query = "http://www.google.cn/music/search?q=#{CGI::escape(ARGV.first)}&aq=f"

target_page = a.get(query)
doc = Nokogiri::HTML(target_page.parser.to_s)

nodes = doc.xpath('//div[@class="results"]/table[@id="song_list"]/tr')
song_ids = []
puts "-- found #{nodes.size} nodes"
log("=--- #{Time.now.strftime('%Y-%m-%d %H:%M')} found #{nodes.size} target node---=")
log("- Starting...")
nodes.each { |song| song_ids << (song.attribute('id').to_s.gsub('row','')) }
song_ids.each_with_index do |sid,index|
  target = "http://www.google.cn/music/top100/musicdownload\?id=#{sid}"
  log("- Download sort : #{index+1}")
  log("- Download page : #{target}")
  down_base = a.get(target)
  down_base.links_with(:href => %r{/music/}).each do |link|
    next if link.href !~ %r{/music/}
    down = link.href.gsub('/music/top100/url?q=','').split('&').first
    file_name = CGI.unescape(down).split('/').last 
    unescape_down = CGI.unescape(down).gsub(/\(/,'\(').gsub(/\)/,'\)')
    unescape_file_name = CGI.unescape(file_name).gsub(/\s/,'-').gsub(/\(/,'').gsub(/\)/,'')
    unescape_file_name = unescape_file_name.gsub(/\'/,'')
    log("- File name : #{unescape_file_name}")
    log("- File url  : #{unescape_down}")
    system("curl -o #{unescape_file_name} #{unescape_down}")
    puts "---> curl -o #{unescape_file_name} #{unescape_down}"
  end
end
log("- Ending...")