geekopardy_challenge_01.rb

# Given a time ordered list of pageviews ("events"), 
# determine how many "visits" each user had, 
# how long each visit lasted and how many page views occurred during each visit. 
# A visit ends when the user has not had another page view for 1 hour, 
# a new pageview after 1 hours starts a new visit (the time between visits is 1 hour).

require 'enumerator'

class Array
  def to_event
    Event.new(self)
  end 
end

class Event < Array
  def user_id
    self[2]
  end 

  def when
    Time.at(self[3].to_i)
  end 

  def valid?
    self.user_id != "NULL"
  end 
end

class UserStadistics < Hash
  def <<(event)
    unless self.keys.include? event.user_id 
      key = event.user_id
      self[key] = { :visits => [] }
      self[key][:visits] << { :count => 1, :begin  => event.when , :end => event.when }
    else
      if ( event.when - self[event.user_id][:visits].last[:end] ) < 3600
        self[event.user_id][:visits].last[:count] = self[event.user_id][:visits].last[:count]  + 1 
        self[event.user_id][:visits].last[:end]  = event.when
      else
        self[event.user_id][:visits] << { :count => 1, :begin  => event.when, :end =>  event.when }
      end 
    end 
  end 

  def get_file_hash(filename)
    events = []
    lines = File.new(filename).readlines
    lines.each do |line|
      row = line.strip.split("\t").to_event
      events << row 
    end 
    events.select{|row| row.valid?}
  end 

  def get_visits(filename)
    events = get_file_hash(filename)
    events.each do |event|
      self << event
    end 
  end 

  def print
    self.each do |user|
      p "-------------------------------------------------------------"
      p "User ID: #{user[0]}"
      p "  Visits: #{user[1][:visits].count} "
      user[1][:visits].enum_for(:each_with_index).collect do |visit, index|
        p "    Visit #{index + 1} - #{visit[:count]} pages views Duration: #{visit[:end] - visit[:begin]} secs"
      end 
    end 
    p "================================================================"
  end 
end

users = UserStadistics.new
users.get_visits('./events.csv')
users.print