console_output

$ git clone git://gist.github.com/582351.git html_entities
...
$ cd html_entities/
$ wget http://www.w3.org/2003/entities/2007xml/unicode.xml
...
$ ruby html_entities.rb > entities.rb
$ ruby html_entities.rb plain > entities.dat
$ ls -sh entities.*
28K entities.dat  68K entities.rb


$ irb

>> load "entities.rb"
=> true
>> HTMLEntities.size
=> 2264
>> puts HTMLEntities["Aacute"]
Á
=> nil

>> entities = File.read("entities.dat").split("\0").inject({}) {|hash, line| line = line.split(" ", 2); hash[line[0]] = line[1]; hash }; entities.size
=> 2264
>> puts entities["euro"]
€
=> nil

html_entities.rb

# Download unicode.xml from http://www.w3.org/2003/entities/2007xml/unicode.xml

require 'rubygems'
require 'nokogiri'

entities = {}
Nokogiri::XML(File.read("unicode.xml")).search("entity").each do |entity|

  entity_name = entity["id"].to_s
  if entity_name =~ /\A\w+\Z/
    if unicode_id = entity.parent["id"][/\AU([0-9a-z]+)\Z/i, 1]
      entities[entity_name] = [unicode_id.to_i(16)].pack('U')
    end
  end
end

case ARGV[0]
when nil, "ruby-hash"

  # Generate a nice ruby hash.
  # It can be load directly in a ruby script

  puts "HTMLEntities = {"
  entities.keys.sort.each do |entity_name|
    puts "  #{entity_name.inspect} => #{entities[entity_name].inspect},"
  end
  puts "}"

when "plain"
  # Plain file. Entities separated by \0. Name and character separated by an space
  # It can be load with
  # ruby> File.read("entities").split("\0").inject({}) {|hash, line| line = line.split(" ", 2); hash[line[0]] = line[1]; hash }
  entities.keys.sort.each do |entity_name|
    print entity_name
    print " "
    print entities[entity_name]
    print "\0"
  end
end