gistfile1.txt

require 'open-uri'
require 'nokogiri'

# スクレイピング先のURL
url = 'http://ishidatozanjukunisshi.blogspot.jp'
#url = 'http://ishidatozanjukunisshi.blogspot.jp/search?updated-max=2010-06-20T21:17:00%2B09:00&max-results=7&start=985&by-date=false'
charset = nil

def url_set(url)
  html = open(url) do |f|
    charset = f.charset # 文字種別を取得
    f.read # htmlを読み込んで変数htmlに渡す
  end
  return html
end

def content_output(url, charset)
  html = url_set(url)
  doc = Nokogiri::HTML.parse(html, nil, charset)
  doc.css('.post-outer').each do |node|
  #  p node
    puts "---------------------"
    puts node.css('.post-title a').text
    puts node.css('.timestamp-link').attribute('href').value
    puts node.css('.timestamp-link > abbr').attribute('title').value
    puts node.css('.post-body').text
  end
end

def content_check(url, charset)
  html = url_set(url)
  doc = Nokogiri::HTML.parse(html, nil, charset)
  if doc.search('.blog-pager > span > .blog-pager-older-link').size == 1 then
    url = doc.css('.blog-pager > span > .blog-pager-older-link').attribute('href').value
  else
    url = ""
  end
  return url
end

#content_output(url, charset)
#content_check(url, charset)
#exit

while url != "" do
  puts url
  content_output(url, charset)
  url = content_check(url, charset)
end