iTS健康診断ページから空き情報一覧を取得する

medical_checkup.rb

require 'open-uri'
require 'nokogiri'
require 'mechanize'

namespace :scrape do
  desc '健康診断ページから空き情報一覧を取得する'
  task :medical_checkup => :environment do
    @courses = {
      # 1 => "1日人間ドック(午前)",
      2 => "健保指定ドック(午前)",
      # 4 => "1日人間ドック(午後)",
      5 => "健保指定ドック(午後)",
    }

    def scrape_yoyaku(index, is_gynecology=false)
      agent = Mechanize.new
      base_url = 'http://www.its-kenpo.or.jp/kanri/chokuei/yoyaku/privacy_yoyaku_kojinyoyku.html'
      page = agent.get(base_url)
      page = page.link_with(:text => '同意する').click
      # p agent.page.title

      # 検索
      page.form_with(:id => 'SearchSpaceIndexForm') do |f|
        f.radiobutton_with(:id => 'SearchSpaceCondExaminationYear2017').check
        f.radiobutton_with(:id => "SearchSpaceCondCourseSingle#{index}").check
        if is_gynecology
          f.checkbox_with(:id => "SearchSpaceCondOption1").check
          f.checkbox_with(:id => "SearchSpaceCondOption3").check
        end
        f.click_button
      end

      gynecology = is_gynecology ? " (婦人科検査)" : ""
      p "#{@courses[index]}#{gynecology}"
      p "dates, place, count"
      elements = agent.page.search('td a')
      elements.each do |e|
        text = e.get_attribute(:onclick) # => selectDate(1, '2018-01-05')
        text = text.gsub(/selectDate\(/, '')
        text = text.gsub(/\)/, '')
        texts = text.split(', ')
        place = texts[0] == "1" ? '大久保健診センター' : '山王健診センター'
        date = texts[1]
        count = e.inner_text

        p "#{date}, #{place}, #{count}"
      end
    end

    @courses.keys.each do |i|
      scrape_yoyaku(i)
      p ""
    end

    @courses.keys.each do |i|
      scrape_yoyaku(i, true)
      p ""
    end

  end
end