application_controller.rb

class ApplicationController < ActionController::Base
  
  after_filter :store_referrer
  
  protected

  def store_referrer
    return if referred_from_here? or request.format != 'text/html'

    referrer_data = {
      :session_id => request.session_options[:id],
      :referral_url => request.referer,
      :requested_url => request.fullpath,
      :user_agent => request.user_agent,
      :referrer_id => params[:ref_id],
      :user_id => current_user.try(:id)
    }

    SiteVisit.delay.create!(referrer_data)
  end
  
  private

  def referred_from_here?
    instance_variable_defined?(:@referred_from_here) ? @referred_from_here : begin
      @referred_from_here = (URI.parse(request.env['HTTP_REFERER']).host == request.host rescue false)
    end
  end
end

site_visit.rb

class SiteVisit
  include Mongoid::Document
  include Mongoid::Timestamps

  field :session_id, :type => String
  field :referrer_id, :type => String

  field :referral_url, :type => String
  field :referral_domain, :type => String
  field :referral_domain_with_subdomain, :type => String

  field :user_agent, :type => String
  field :requested_url, :type => String

  field :is_bot, :type => Boolean

  belongs_to :user, index: true

  index session_id: 1
  index referrer_id: 1

  before_create :sanitize
  before_create :save_referer_domain, :save_is_bot

  private

  def sanitize
    # this quietly replaces non-UTF-8 characters with "?"
    self.user_agent = user_agent.try :encode, 'UTF-8', undef: :replace, replace: '?'
  end

  def save_referer_domain
    return if self.referral_url.blank?

    url = self.referral_url
    url = "http://#{url}" unless URI.parse(url).scheme

    parsed_url = URI.parse URI.escape url

    self.referral_domain_with_subdomain = parsed_url.host

    if parsed_url.host
      host = parsed_url.host.downcase
      self.referral_domain = host.split('.').last(2).join('.')
    end
  rescue
    # ignore
  end

  def save_is_bot
    if self.user_agent.blank? or self.user_agent.match(/asynchttpclient|alexa|bot|butterfly\/\d.\d|crawl(er|ing)|crowsnest|curl|embedly|eventmachine|facebookexternalhit|feedburner|flipboardproxy|firephp|google web preview|(^java)|LongURL|MetaURI|nagios|(^NING\/\d.\d$)|news\.me|peerindex|pingdom|postrank|(^python)|rockmeltembedservice|(^ruby$)|slurp|spider|viadeo|yahoo!|yandex/i)
      self.is_bot = true
    else
      self.is_bot = false
    end
    true
  end
end