data_get_aeds_rake.rb

  desc "Get AED Places from web pages"
  task :get_aeds, [:begin_n, :thru_n] => :environment do | t, args |
    args.with_defaults(:begin_n => "1", :thru_n => "1277") # [..) 1277
    type = :aed
    Place.type(type).delete
    begin_n = args.begin_n.to_i; thru_n = args.thru_n.to_i
    n = begin_n
    begin
      cnt = 0
      puts "EGEN,FETCH,#{n}"
      # fetch list
      aed_places = egen_get_list 
      unless aed_places.nil? 
        aed_places.each do |el|
          aed = el['href']
          unless aed.empty?
            cnt = cnt + 1
            aed = aed.split("'")
            name = aed[1].gsub(/,/,'_')
            sn = aed[9]
            begin
              info = {}
              # jobs
              info[:category], address, phone, info[:address_desc], nop, info[:created_at], info[:model] = egen_print_info sn
              info[:sn] = sn
              print type, ",", name, ",", address[13..-1], ",", phone, ",", info, "\n" if Rake.application.options.trace == true
              Place.create!(
                type: type,
                name: name,
                address: address,
                phone: phone,
                info: info
              )
            rescue OpenURI::HTTPError
              puts "ERROR,#{n},#{cnt},NOT_FOUND,-"
              next
            end
          end
          sleep 0.42
        end #aed_places
      end
    end
  end

database_helper.rb

module DatabaseHelper

require 'net/http'
require 'nokogiri'
require 'open-uri'
require 'timeout'

TIMEOUT_CNT = 42

  def egen_print_info code
    retries = TIMEOUT_CNT
    begin
      Timeout::timeout(5) {
        ret = []
        doc = Nokogiri::HTML(open("http://www.e-gen.or.kr/egen/inf.AED2.do?yearSeq=#{code}?HPID="))
        doc.xpath('//ul/li').each do |a|
          header, data = a.text.gsub(/[\t]/, '').gsub(/\r\n/, ' ').gsub(/,/,'_').split(': ')
          ret.push(data)
        end
        ret
      }
    rescue Timeout::Error
      retries -= 1
      if retries > 0 
        puts "sleep"
        sleep 1.42
        retry
      else
        puts "raise"
        raise
      end
    end
  end # def print_info

  def egen_get_list n
    retries = TIMEOUT_CNT
    elements = nil
    begin
      cnt = 0
      Timeout::timeout(5) {
        doc = Nokogiri::HTML(open("http://www.e-gen.or.kr/egen/inf.AED1.do?lon=&lat=&cnt=2444&str_cnt=0&page_num=1&page_size=100000&radius=400000&x=20&y=15&page=#{n}"))
        elements = doc.xpath('//td/b/a')
      }
    rescue Timeout::Error
      retries -= 1
      if retries > 0 
        puts "sleep"
        sleep 1.42
        retry
      else
        puts "raise"
        raise
      end
    end
    elements
  end
end

geocode_helper.rb

module GeocodeHelper

  DAUM_APIKEY = YAML.load_file(Rails.root.join('config','apikey.yml'))['daum']

  def geocode(address)
    url = "http://apis.daum.net/local/geo/addr2coord?apikey=#{DAUM_APIKEY}&output=json&q=#{address}"
    items = JSON.parse(open(URI.encode(url)).read)['channel']['item']
    unless items.blank?
      item = items.first
      [item['lng'], item['lat']]
    end
  end

end

Place.rb

class Place
  include Mongoid::Document
  include Mongoid::Timestamps
  include Geocoder::Model::Mongoid

  field :type
  field :category
  field :category_desc
  field :name
  field :description
  field :info, type: Hash

  field :zipcode
  field :address
  field :coordinates, type: Array
  field :phone

  geocoded_by :address
  after_validation :geocode

  index({ coordinates: "2d" })

  # scope
  def self.type(type)
    where(type: type)
  end
end