Grabs all public photos from a friendster account ( www.friendster.com )

friendster_photo_grab.rb

require 'rubygems'
require 'nokogiri'
require 'open-uri'
require 'json'
require "cgi"
require 'fileutils'

def fdoc(rel_url)
  Nokogiri::HTML(open("http://www.friendster.com#{rel_url}"))
end

def user_name(id = USER_ID)
  @user_name ||= fdoc("/#{id}").css(".username").first.content.to_s.split(' ').map(&:capitalize).join(' ')
end

#
#configuration
#
USER_ID = 1234567

DEST_DIR = "#{USER_ID} - #{user_name}"

puts "User: #{USER_ID} (#{user_name})"


def album_links(user_id = USER_ID)
  links = []

  page_nr = 0
  while true
    page = fdoc("/viewalbums.php?page=#{page_nr}&uid=#{user_id}")
    new_links = page.css('a.albumThumb')

    break if new_links.empty? # break out of the loop if no albums found anymore

    links = links + new_links
    page_nr = page_nr + 1
  end

  puts "User has #{links.length} Album(s)"

  links
end

def picture_json_strings(album_id, user_id = USER_ID)
  json_strings = []

  page_nr = 0
  while true
    page = fdoc("/viewphotos.php?page=#{page_nr}&a=#{album_id}&uid=#{user_id}")

    break if page.css("div#photoGallery a").empty?

    gallery_url = page.css("div#photoGallery a").first.attributes['href']
    gallery_site = fdoc(gallery_url)

    script_tag = gallery_site.css("div#twoColContainer script").first
    js_line = script_tag.content.split("\n").select {|l| l.include?('_pl')}.first
    json = js_line.split('=', 2).last.strip.chop

    json_strings << json

    page_nr = page_nr + 1
  end

  puts "Album has #{json_strings.length} page(s)"
  json_strings
end

def sanitize_file_name(string)
  string.gsub(/[^0-9A-Za-z.\-]/, ' ').gsub(/\s+/, ' ').strip
end


# Go through every album page
album_links.each do |album_link|
  album_title = album_link.attributes['title'].to_s
  album_url = album_link.attributes['href'].to_s

  puts ""

  # Skip private albums
  if album_url.include?("private")
    puts "Skipping Album: #{album_title} (private)"
    next
  end
  puts "Album: '#{album_title}'"
  
  #get Album ID from URL
  album_id = CGI::parse(album_url.split('?').last)['a']

  dir_name = sanitize_file_name(album_title)
  dir_path = File.join("#{DEST_DIR}", dir_name)
  FileUtils.mkdir_p(dir_path) if not File.exists?(dir_path)

  dir_entries = Dir.entries(dir_path)

  picture_json_strings(album_id).each_with_index do |json, photo_page_nr|
    puts "Downloading pictures from page #{photo_page_nr}"

    photos = JSON.parse(json)['photos']
    photos.each_with_index do |photo, photo_nr|
      photo_url = photo['url']
      photo_caption = photo['caption']
      photo_caption = sanitize_file_name(photo_caption)

      photo_number = '%03d' % (photo_page_nr * 20 + photo_nr)
      photo_name = photo_number
      photo_name = "#{photo_number} #{photo_caption}" unless photo_caption.empty?
      
      file_base_name = "#{photo_name}.jpg"
      file_name = File.join(dir_path, file_base_name)

      # Skip if file exists
      if File.exist?(file_name)
        puts "Skip: #{file_base_name} exists"
        next
      end

      # Rename file with updated caption (only based on self generated photo number)
      file_found = dir_entries.select {|f| File.basename(f).index(photo_number) == 0 }
      unless file_found.empty? or file_found.first == file_base_name
        file_found = File.join(dir_path, file_found.first)
        puts "Rename: #{File.basename(file_found.first)} to #{file_base_name}"
        FileUtils.mv(file_found, file_name)
        next     
      end

      # Download file
      begin
        File.open(file_name, 'wb') do |file|
          puts "Download: #{File.basename(photo_url)} as #{File.basename(file_name)}"
          file.write(open(photo_url).read)
          sleep 0.5
        end
      rescue
        puts "ERROR saving #{photo_url}"
      end
      
    end
  end

end

puts "FINISHED!"