Systematically retrieve audio files from popular mp3 blogs

mp3_update

#!/usr/bin/env ruby
require 'nokogiri'
require 'open-uri'

class Mp3Crawler
  def initialize(url)
    @data = Nokogiri::XML(open(url))
  end

  def links
    @data.xpath('//enclosure').select{|e| e.attributes['type'].value.strip == 'audio/mpeg' }.map do |e|
      e.attributes['url'].value.strip
    end
  end
end

feeds = DATA.read.split("\n").reject(&:empty?)

links = feeds.reduce([]) do |links, feed|
  puts "Crawling #{feed}"
  links << Mp3Crawler.new(feed).links
  links
end

links.flatten!

puts "Captured #{links.size} files. Starting download..."

soundcloud = ->(l) { l =~ /soundcloud\.com/ }

links.each do |link|
  filename = URI.decode(File.basename(case link
                                      when soundcloud then link[/\/(.*)\/download/n, 1]
                                      else link
                                      end))

  filename.concat('.mp3') if File.extname(filename).empty?
  path = File.join(ENV['HOME'], 'Music', 'Blogs', filename)

  puts "Downloading #{link}"

  `wget -cO '#{path}' #{link}`
end

__END__

http://feeds2.feedburner.com/aurgasm
http://feeds.feedburner.com/tinyways
http://boomboomchik.com/feed
http://www.audiodrums.com/feed/
http://www.qcmixtapes.com/?feed=rss2
http://bambooorchestra.blogspot.com/feeds/posts/default?alt=rss
http://www.undomondo.com/rss
http://www.mixtaperiot.com/feed/
http://www.blogotheque.net/feed/
http://feeds2.feedburner.com/bigstereo
http://palmsout.net/feed/
http://www.scissorkick.com/feed/
http://music.for-robots.com/?feed=rss2
http://soul-sides.com/feed/
http://bennloxo.com/feed/
http://3hive.com/feed/