require 'open-uri'
require 'nokogiri'

SLEEP_TIME = 3

root_url = 'http://www.capcom.co.jp/arcade/rev/PC/ranking_highscore.html'
page_url_base = root_url
detail_url_base = 'http://www.capcom.co.jp/arcade/rev/PC'
charset = nil

DIFFICULTIES = {
  '../common/img_common/bnr_difficulty_easy.png' => 'esy',
  '../common/img_common/bnr_difficulty_standard.png' => 'std',
  '../common/img_common/bnr_difficulty_hard.png' => 'hrd',
  '../common/img_common/bnr_difficulty_master.png' => 'mas',
  '../common/img_common/bnr_difficulty_unlimited.png' => 'unl',
}

def parse_page(url)
  charset = nil
  begin
    html = open(url) do |f|
      charset = f.charset
      f.read
    end
    sleep SLEEP_TIME
    Nokogiri::HTML.parse(html, nil, charset)
  rescue
    sleep 5
    retry
  end
end

# --begin

doc = parse_page(root_url)
page_urls = []
doc.xpath('//div[@class="mPager-top"]/a[@class="pageing" or @class="pageing2"]').each do |node|
  page_urls << "#{page_url_base}#{node.attribute('href').value}"
end
page_urls.uniq!

page_urls.each do |page_url|
  page_doc = parse_page(page_url)
  page_doc.xpath('//ul[@class="rkLinkList"]/li/a').each do |node|
    detail_doc = parse_page(node.attribute('href').value.gsub(/\A\./, detail_url_base))
    container = []
    container << File.basename(detail_doc.xpath('//div[@class="rkHiscoreMusicCv"]/img').attribute('src').value)
    container << DIFFICULTIES[detail_doc.xpath('//div[@class="rkHiscoreDiffBlock"]/img').attribute('src').value]
    detail_doc.xpath('//div[@class="rkHiScoreDetail"]').each do |record|
      container << record.css('p.rankScore').text.to_i
    end
    STDERR.puts container[0, 2].join("\t")
    puts container.join("\t")
  end
end