Script to explore chess opening lines based on most popular google results

goopening.rb

#!/usr/bin/env ruby

#Script to explore chess opening lines based on most popular of google's top 100 results
#Usage:
#▸ ./goopening "1. e4 e5 2."
#1: nf3 (35)
#2: qh5 (7)
#3: bc4 (4)
#4: nc6 (2)

require 'open-uri'
require 'nokogiri'
require 'cgi'
require 'stringio'

def normalize(alg)
  alg = alg.gsub(/\.\S/, '. ')
  alg = alg.gsub(/(\d)\./, '\1')
  alg.downcase
end

def find_chess_move(dirty_string)
  move_regexp = /([kqnpbr](([abcdefgh]?[1-8]?))?x?[abcdefgh][1-8]|o-o(-o)?|ooo?)/
  dirty_string.scan(move_regexp).flatten.first
end

def get_strings_at_the_right_of_each_match(text_of_search_page, move_list)
  size_of_algebraic_notation_movement = 15
  text_of_search_page.scan(/#{move_list}(.{#{size_of_algebraic_notation_movement}})/).flatten
end

move_list = ARGV[0]
move_list = normalize(move_list)
text_of_search_page = Nokogiri(open("http://www.google.com/search?q=#{CGI.escape(move_list)}&num=100&complete=0").read).text
text_of_search_page = normalize(text_of_search_page)

strings_at_the_right_of_each_match = get_strings_at_the_right_of_each_match(text_of_search_page, move_list)

count_hash = {}
strings_at_the_right_of_each_match.map{|a| find_chess_move(a)}.compact.map do |move|
  count_hash[move] = 0 unless count_hash[move]
  count_hash[move] += 1
end

best = count_hash.to_a.sort{|a, b| b.last <=> a.last }
i = 0
puts best.map{|b| b.last == 1 ? nil : "#{i+=1}: #{b.first} (#{b.last})"}.compact