CSA形式の将棋の棋譜の完全一致チェック．kifu_check.rbの最後の部分でパラメーターを適宜編集．lib/csa_reacord.rbとlib/tree.rbを配置して利用．

csa_reacord.rb

# -*- coding: utf-8 -*-
# CSA形式の棋譜を表すクラス

class CsaRecord
  Sente = true
  Gote  = false
  attr_accessor :file
  def initialize( *args )
    option = args.pop
    case option
    when File
      @file = option
    when String
      @file = option
    else
      raise "This initialize method can accept only File or String obj."
    end
    @record = Array.new
    @sente_time = Array.new
    @gote_time = Array.new
    @turn = Sente
    @ply  = 0
    @initial_position = "PI"
    @description = ""
    @result = Hash.new
    self.parse()
  end

  def flip_turn()
    @turn = @turn ? Gote : Sente
  end

  def parse()
    (@file.kind_of? File) ?
    self.parse_csa_string(@file.read) : self.parse_csa_string(@file)
  end

  def parse_csa_string(str)
    str.each_line do |line|
      self.parse_csa_line(line.chomp)
    end    
    @result
  end

  def [](num)
    @record[num]
  end

  def each()
    @record.each do |move|
      yield move
    end
  end

  def length
    @record.length
  end

  def parse_csa_line(line)
    case line
    when /^N\+(.+)/ 
     @sente_name = $1
    when /^N-(.+)/
      @gote_name = $1
    when  /^\+$/
      @turn = Sente
    when  /^\-$/
      @turn = Gote
    when /^([\+-]\d\d\d\d[A-Z][A-Z])/ 
      @record << $1      
      @ply += 1
      flip_turn()
    when /^PI/
      @initial_position = "PI"
    when /^P.*/
      # PI以外での初期局面
    when /^T.+/ 
      #持ち時間
    when /summary/
      line =~ /summary:.*:.*\s(.*):.*\s(.*)/
      #'summary:toryo:bona1 lose:bona2 win'
      @result[:sente] = $1
      @result[:gote]  = $2
    when /^('.*)/ 
      #コメント
      @description << $1 + "\n"
    when /^%(.+)/ 
      @result[:reason] = $1
    end
  end
end

kifu_check.rb

# -*- coding: utf-8 -*-
# 棋譜の完全重複チェックツール
#
$: << File.dirname(__FILE__)
require 'lib/csa_record'
require 'lib/tree'
require 'pp'


raise "棋譜を含むディレクトリを指定してください!" if ARGV[0].nil?
$kifu_dir = ARGV[0] 

start = Time.now
records = Array.new
Dir.chdir($kifu_dir)
Dir.glob("*.csa") do |csa|
  open(csa, "r") do |f|
    csa = CsaRecord.new(f)
    records << csa
  end
end
puts Time.now-start 
puts "インスタンス化完了!(#{records.length})"

start = Time.now
tree = CsaRecord::Tree.new
records.each do |csa|
  for i in 0..(csa.length)
    ptree = tree
    for j in 0..i
      ptree[csa[j]] = CsaRecord::Tree.new if ptree[csa[j]].nil?
      ptree = ptree[csa[j]]
      ptree.value << csa.file.path if i == j
    end
  end
end

proc_print = Proc.new {  |node, key, ply|
  print "#{ply}:#{key} => "
  p node.value
}

tree.traversal do |node, key, ply|
  if ply > 30 && node.value.length > 1
    puts "棋譜の重複！ #{node.value.inspect}"
  end
end
puts Time.now-start
puts "棋譜の操作終了"

tree.rb

# -*- coding: utf-8 -*-
#複数のCSA形式の棋譜を木構造として構築するための

class CsaRecord
  class Tree
    attr_accessor :child, :value
    def initialize
      @value = Array.new
      @child = Hash.new
    end

    def [](str)
      @child[str]
    end

    def []=(str, val)
      @child[str] = val
    end

    def traversal(key = "root", ply = 0, &func)
      unless self.child.nil?
        child.each do |key, val|
          val.traversal(key, ply + 1, &func)
        end
      end
      func.call(self,key, ply)
    end
  end
end