Contents

#! /usr/bin/env ruby

require 'mechanize'
require 'csv'
require 'swot'
require './lib/gman'
require './lib/gman/parser'

url = "http://www.myndighetsregistret.scb.se/Myndighet.aspx"
agent = Mechanize.new
page = agent.get(url)
form = page.form("form1")
form.radiobuttons.find { |r| r.value = "Textfil" }.check
submit_button = form.buttons.find { |b| b.type == "submit" }
response = agent.submit(form, submit_button)

domains = []
rows = CSV.parse(response.content, :headers => true, :col_sep => "\t")

puts "Starting with #{rows.count} domains..."

rows.each do |row|
  next if Swot.valid?(row["Webbadress"]) # Filter out Swot'd domains
  next if row["Namn"] =~ /UNIVERSITET/ # Filter out domains that are clearly edu
  domains.push Gman.new(row["Webbadress"]).domain.to_s.gsub(/^www\./,"")
end

domains.reject! { |domain| domain.empty? }
domains.compact!
domains.uniq!
domains.select! { |domain| PublicSuffix.valid?(".#{domain}") }

puts "Ended up with #{domains.count} domains."

current = Gman::Parser.file_to_array( Gman::list_path )
current_hash = Gman::Parser.array_to_hash(current)

current_hash["Swedish Administrative Authorities"] = domains
current_hash = current_hash.sort_by { |group, domains| group.downcase }

# PublicSuffix Formatted Output
current_group = ""
output = ""
current_hash.each do |group, domains|
  if group != current_group
    output << "\n\n" unless current_group.empty? # first entry
    output << "// #{group}\n"
    current_group = group
  end
  output << domains.join("\n")
end

File.open(Gman.list_path, "w") { |file| file.write output }

Version data entries

5 entries across 5 versions & 1 rubygems

Version	Path
gman-4.6.5	script/vendor-se
gman-4.6.4	script/vendor-se
gman-4.6.3	script/vendor-se
gman-4.6.2	script/vendor-se
gman-4.6.1	script/vendor-se