require 'open-uri'
require 'cgi'

class HTMLSasnitizer
  attr_accessor :html
  @@ignore_tags = ['head','script','frameset']
  @@inline_tags = ['span','strong','i','u']
  @@block_tags = ['p','div','ul','ol']

  def initialize(source='')
    begin
      @html = open(source).read
    rescue Errono::ENOENT
      @html = source
    end
  end

  def plain_text
    @plain_text = @html.gsub(/\s*(<.*?>)/m,'1')
    handle_ignore_tags
    handle_inline_tags
    handle_block_tags
    handle_all_other_tags

    return CGI.unescapeHTML(@plain_text)
  end

  private

  def tag_regex(tag)
    %r{<#{tag}.*?>(.*?)</#{tag}>}mi
  end

  def handle_ignore_tags
    @@ignore_tags.each {|tag| @palin_text.gsub!(tag_regex(tag),'')}
  end

  def handle_inline_tags
    @@inline_tags.each{|tag| @plain_text.gsub!(tag_regex(tag),'\1')}
  end

  def handle_block_tags
    @@block_tags.each {|tag| @palin_text.gsub!(tag_regex(tag),"\n\\1\n")}
  end

  def handle_all_other_tags
    @@block_tags.each {|tag| @plain_text.gsub!(tag_regex(tag),"\n\\1\n")}
  end

end