#!/usr/bin/ruby -Ku
# encoding: utf-8
require 'amalgalite'
require 'fileutils'

class Csv_to_db
  def initialize(db_name, csv_name)
    db_file = "#{db_name}.db"
    FileUtils.rm(db_file) if File.exist?(db_file)  # start with a clean slate
    my_db=Amalgalite::Database.new(db_file)
    csv_rows = File.read("#{csv_name}", :encoding => "utf-8").split("\n")
    col_names = csv_rows[0].split(',')
    col_names_types= col_names.map do |col| col + ' char(1)' end
    sql_create = "create table ctr_map(%s)" % col_names_types.join(', ')
    my_db.execute(sql_create)
    stmt = csv_rows[1..-1].map do |x|
      "insert into ctr_map (%s) values (%s)" % [col_names.join(', '), x]
    end
    stmt.each do |row_insert|
      my_db.execute(row_insert)
    end
    puts "ok"
  end
end

db_name = "ctr_map"
csv_name = "chinese_map.csv"
Csv_to_db.new(db_name,csv_name)

  def save_to_db
    str = self.gsub(/,/,"\",\"").gsub(/^/,"\"").gsub(/$/,"\"")
    my_db=Amalgalite::Database.new(@@gsp_db)
    stmt = my_db.execute("select * from gsp_corpus where word = ?", "#{self.split(',')[0]}")
  if !(stmt.any?)
   insert_stmt = "insert into gsp_corpus (%s) values (%s)" % [@@field_name.split.join(','), str]
  end
    my_db.execute(insert_stmt)
    my_db.close
  end

#str = "教育部,gs"

#!/usr/bin/ruby -Ku
# encoding: UTF-8
#將停用字移除
class String
  def to_rm_stop_words(stop_words)
    msg = self
      stop_words.each do |e|
        msg = msg.gsub("#{e}",',')
      end
    return msg
  end
end

stop_words_db = "stop_words.csv" #stop words list
dst_dir = "dst_dir" #處理檔案放置位置
file_array = Dir.glob("*.txt")

#stop_words
stop_words = Array.new
  File.open("#{stop_words_db}") do |content|
    while line = content.gets
      stop_words << line.strip
    end
  end

file_array.each do |each_file|
  msg = String.new #msg的型別為string
  File.open("#{each_file}") do |content|
    while line = content.gets
      msg << line.to_s.strip.chomp
    end
  end
   revised_msg = msg.to_rm_stop_words(stop_words).gsub(/,+/,",")


  File.open("./#{dst_dir}/#{each_file}","a") do |content|
    content.puts revised_msg
  end
  msg.clear
end

class String

  def mbsubstr1(*range)

      split(//)[*range].join("")

  end

   def mbsubstr2(idx, len)

       slice(/\A.{#{idx}}(.{0,#{len}}/m, 1)

   end

end

p "一大串字一大串字".mbsubstr1(2,3)

p "一大串字一大串字".mbsubstr1(2..3)

p "一大串字一大串字".mbsubstr2(2,2)

#str 一段文字, term 字典的中的詞

def mmseg(str,term)
strEnd = str.length-1
  loop do
    tmpStr=str.substr(0..strEnd)
    strEnd -= 1
    sleep 0.1
    puts tmpStr

    if tmpStr == term
      print "I found #{tmpStr}"
      break
    end
    if (strEnd == -1)
      str=str.substr(1..-1)
      strEnd = str.length-1
      if str.length == 1
        puts "no match"
        break
      end
    end
  end
end

 f = { 1 => ->(x) { puts "#{x}  from 1" },

         2 => ->(x) { puts "#{x} bye from 2" }

      }

f.call[1]("mary")

n=4
i=1
j=i+1

while (i<4) do
  while (j<=n) do
    printf("%d,%d\n",i,j)
    j += 1
  end
  i += 1
  j = i+1
end

require 'ai4r'

data = [[1,1,1], [3,3,3],[9,9,9]]
ai4r_data = Ai4r::Data::DataSet.new(:data_items=> data)

kmeans = Ai4r::Clusterers::KMeans.new

#abuild(dataset,clusters)

res=kmeans.build(ai4r_data,2)

puts res.inspect

puts res.centroids.join(',')

dis = Ai4r::Clusterers::KMeans.new

a = [1,1]
b = [3,3]

puts Math.sqrt(dis.distance(a,b))