For more detail: http://june29.jp/2010/12/17/glamorous-rubysappororecipe/

downloader.rb

images = "images"
Dir.mkdir images unless File.exist? images

horesasu = Jigokuno::Misawa.new
horesasu.each { |meigen|
  open("%s/%s.gif" % [images, meigen], "w") { |gif|
    gif.puts open(meigen.image).read
  }

  sleep 5
}

jigokuno.rb

# -*- coding: utf-8 -*-

require "open-uri"

require "rubygems"
require "nokogiri"

module Jigokuno
  URL        = "http://jigokuno.com/"
  NEXT_XPATH = "//div[@id='page_area']/div[@class='page_navi']/a[contains(text(), '>>')]"

  class Misawa
    include Enumerable

    attr_reader :current

    def initialize
      @current  = URL
      @document = Nokogiri::HTML(open(@current))
    end

    def next
      next_link = @document.at(NEXT_XPATH)

      return nil if next_link.nil?

      @current  = next_link["href"]
      @document = Nokogiri::HTML(open(@current))

      self
    end

    def each
      Scraper.new(@document).meigens { |meigen| yield meigen }

      next_page = self.next
      return if next_page.nil?

      next_page.each { |meigen| yield meigen }
    end
  end

  class Scraper
    def initialize(html)
      @html = html
    end

    def meigens
      @html.xpath("//div[@class='entry_area']").each { |entry|
        h2 = entry.xpath("h2").text.tr("０-９", "0-9").chomp
        id, title = h2.scan(/惚れさせ(\d+).*「(.+)」/).first

        id        = id.to_i
        title     = title.to_s
        image     = entry.at(".//img[@class='pict']").attributes["src"].to_s
        character = entry.at("center/ul/li[2]/a").text

        yield Meigen.new(id, title, character, image)
      }
    end
  end

  class Meigen
    attr_reader :id, :title, :character, :image

    def initialize(id, title, character, image)
      @id        = id
      @title     = title
      @character = character
      @image     = image
    end

    def to_s
      "%03d_%s_%s" % [@id, @title, @character]
    end
  end
end