analytics_wikipedia_access_data_older_version.rb

# encoding: utf-8
# Analyze Wikipedia access data older version
# https://dumps.wikimedia.org/other/pagecounts-raw/
require "cgi"
filename = "20120301-000000-ja.txt"
file = File.open(filename, "r:UTF-8")
list = []
while text = file.gets
  begin
    next unless text =~ /^ja/
    data = text.split
    h = {:title => CGI.unescape(data[1]), :count => data[-2]}
    list.push h
  rescue Exception => e
    #p e
  end
end
file.close

# count順にソート
result = list.sort_by do |i|
  i[:count].to_i
end

# トップ20表示
result.reverse.first(20).each do |i|
  puts i
end