Exports every wall post from vkontakte.ru to a CSV Updated in Feb 2013, Ruby 1.9

get.rb

# Usage:

# 1. Install required gems
# $ sudo gem install mechanize progressbar
#
# 2. Run with [vk.com|vkontakte.ru] [remixsid cookie] and [your_profile_id]
# $ ruby ./fetch.rb vk.com a528621366f13fb859a120b3a8c357dc98c1f7d93f5e66666662 8666676
# 54% |ooooooooooooooooooooooooooooooooooooooooooooooooooooo                                              | ETA:   0:01:49
#
# 3. Enjoy the output
# $ cat ./export_vk_wall.csv


#!/usr/bin/env ruby
require 'rubygems'
require 'mechanize'
require "csv"
require 'progressbar'

sitename, cookie, gid = ARGV
url = URI.parse("http://#{sitename}/wall#{gid}")

agent = Mechanize.new
agent.user_agent_alias = 'Mac Safari'
Mechanize::Cookie.parse(url, "remixsid=" + cookie) { |c| agent.cookie_jar.add(url, c) }

pages = agent.get(url.to_s).search('#fw_summary_wrap .pg_lnk:last').attr('href').value.split('=').last.to_i/20

pbar = ProgressBar.new("test", pages * 20)

CSV.open('export_vk_wall.csv','w') do |csv|
  csv << ['name', 'message', 'time']
  (0...pages).to_a.each do |p|
    url.query = "offset=#{p * 20}"
    page = agent.get(url.to_s).search("#page_wall_posts")
    page.search('.post_info').each do |post|
      pbar.inc
      csv << [
        post.search('a.author').text,
        post.search('.wall_post_text').text,
        post.search('.rel_date').text
      ]
    end
    sleep(5 + rand(11))
  end
end  
pbar.finish