linkgrabber.rb

#tweetから記事リンクがあるものだけを取得する。

#coding utf-8

require 'twitter'
require 'net/http'
require 'uri'

class LinkGrabber
  def initialize(user)
    @user = user
  end

  def timeline
    Twitter.user_timeline(@user)
  end

  def expand_url(url)
    begin
      uri = url.kind_of?(URI) ? url : URI.parse(url)
    rescue URI::InvalidURIError => e
      puts e
      return nil
    end

    begin
      res = Net::HTTP.start(uri.host, uri.port) do |io|
        io.head(uri.path)['Location']
      end
    rescue EOFError
      return uri.to_s
    rescue Errno::ECONNRESET => e
      puts e
      retry
    end
    
    return uri.to_s if res == nil
    expand_url(res)
  end

  def links
    links = []
    texts = timeline.map {|item| item['text']}

    texts.each do |text|
      links << expand_url($1) if text =~ %r{(http://.*?)(\s|\z|[^A-Za-z0-9./#?_-])}
    end
    links
  end
end

if __FILE__ == $0
  lg = LinkGrabber.new('user_name')
  p lg.links
end