National Gallery scraper

require 'builder'
require 'nokogiri'

BASE_URL = 'http://www.nga.gov'
FEED_URL = BASE_URL + '/programs/calendar/'

def national_gallery_of_art(page)
  doc = Nokogiri::HTML page

  xml = Builder::XmlMarkup.new :indent => 2
  xml.instruct! :xml, :version => 1.0, :encoding => 'utf-8'

  xml.feed(:xmlns => 'http://www.w3.org/2005/Atom') do |feed|
    feed.title 'National Gallery of Art'
    feed.link :href => FEED_URL
    feed.author do |author|
      author.name 'National Gallery of Art'
    end

    # feed.updated
    # feed.id ?

    #each anchor corresponds to a day
    anchors = doc.xpath("//a[starts-with(@id, 'date')]").to_a.map {|a| a['id']}

    anchors.each_index do |i|
      path = <<-XPATH
      //div[@class='event-grouping'
            and preceding-sibling::a[@id='#{anchors[i]}']][1]
      XPATH

      day = doc.xpath(path).text

      # select all events between two consecutive anchors
      path = "//div[@class='coe-list' and preceding-sibling::a[@id='#{anchors[i]}'] "

      if i < anchors.length - 1
        path << " and following-sibling::a[@id='#{anchors[i+1]}']"
      end

      path << "]"

      events = doc.xpath path
      time = events.first.xpath("span[@class='time'][1]").text.strip

      events.each do |event|
        new_time = event.xpath("span[@class='time']").first
        if new_time
          time = new_time.text.strip
        end

        div = event.xpath("div[@class='event'][1]")
        title = div.text
        link = BASE_URL + div.xpath('a[1]').attr('href')

        feed.entry do |entry|
          entry.title title, :type => 'text'
          entry.link link
          entry.summary "#{day}, #{time}", :type => 'text'
        end
      end
    end

  end # xml.feed

  xml.target!
end