Quick hack to parse Folkets Lexikon xml dump into word pairs to import into Anki

parse.rb

# http://folkets-lexikon.csc.kth.se/folkets/om.en.html
# http://folkets-lexikon.csc.kth.se/folkets/folkets_en_sv_public.xml
# http://folkets-lexikon.csc.kth.se/folkets/folkets_sv_en_public.xml

require "rexml/document"
require 'cgi'

file = File.new("folkets_en_sv_public.xml")
doc = REXML::Document.new file

File.open('folkets_sv_en.txt', 'w') do |file|
  doc.elements.each("*/word") do |element|
    value = element.attributes["value"]
    translations = []
    element.elements.each("translation") do |child|
      # double unescape because stuff isn't escaped properly
      translations << CGI.unescapeHTML(CGI.unescapeHTML(child.attributes["value"]))
    end
    # filter out anything without a translation
    if translations.length > 0 then
      file.puts value + "\t" + translations.join(" / ")
    end
  end
end