短縮URLを展開するearthquake.gem用プラグイン。…短縮じゃないやつでもひたすら展開するけど。

deep_expand_url.rb

# -*- coding: utf-8 -*-
# earthquake.gem plugin
# expand url
require 'open-uri'
require 'uri'
require 'net/http'

def get_location(url)
  begin
    url = URI.parse(url.to_s)
  rescue URI::InvalidURIError
    url = URI.parse(URI.escape(url.to_s))
  end

  proxy = url.find_proxy
  if proxy
    clazz = Net::HTTP::Proxy(proxy.host, proxy.port, proxy.user, proxy.password)
  else
    clazz = Net::HTTP
  end

  http = clazz.new(url.host, url.port)
  if url.class == URI::HTTPS
    require 'net/https'
    http.use_ssl = true
    http.verify_mode = OpenSSL::SSL::VERIFY_PEER
    http.cert_store = OpenSSL::X509::Store.new.set_default_paths
  end
  http.open_timeout = 3
  http.read_timeout = 3

  begin
    http.start do
      http.request_head url.request_uri do |res|
        case res
        when Net::HTTPMovedPermanently, # 301
             Net::HTTPFound, # 302
             Net::HTTPSeeOther, # 303
             Net::HTTPTemporaryRedirect # 307
          begin
            location = URI.parse(res['location'])
          rescue URI::InvalidURIError
            return nil
          end
          location = url + location if location.relative?
          return location.to_s
        end
      end
    end
  rescue TimeoutError
    puts "expand timeout: #{url.to_s}" if config[:debug]
  end
  nil
end

def expand_url(url)
  while true
    location = get_location(url)
    break if location.nil?
    puts "expanded: #{url} => #{location}" if config[:debug]
    url = location
  end
  url
end

Earthquake.init do
  output_filter do |item|
    next unless item['text']
    text = item['text']

    entities = (item["retweeted_status"] && item["truncated"]) ?
        item["retweeted_status"]["entities"] : item["entities"]
    if entities
      entities.values_at("urls", "media").flatten.compact.each do |entity|
        url, expanded_url = entity.values_at("url", "expanded_url")
        if url && expanded_url
          deep_expanded_url = expand_url(expanded_url)
          text.sub!(url, expanded_url)
          text.sub!(expanded_url, deep_expanded_url)
        end
      end
    else
      URI.extract(text.dup, %w[http https]) do |url|
        expanded_url = expand_url(url)
        text.sub!(url, expanded_url)
      end
    end
    true
  end
end