libro news page parser

gistfile1.rb

require 'rubygems'
require 'nokogiri'
require 'open-uri'
require 'jcode'

$KCODE='u'
page_content = Struct.new('PageContent', :category, :title, :body, :updated_at)
content = Nokogiri::HTML.parse(open('http://www.libro.jp/news/').read)
contents = content.xpath("//div[@class='entry']").map do |entry|
  page = page_content.new(
    entry.xpath("p[@class='category_txt']").text.to_s,
    entry.xpath("h2").text.to_s.gsub(/[\s　]+/,' '),
    entry.xpath("div[@class='entry_body']/div[@class='text']").text.to_s.gsub(/[\s　]+/,' ').gsub(/<br>/,"\n"),
    Time.now
  )
  day = /([0-9]+)月([0-9]+)日/.match(page.title.tr('０-９','0-9')).to_a
  page.updated_at = Time.local(Time.now.year, day[1], day[2])
  page
end

puts contents