# -*- coding: utf-8 -*-
# @ description:
# @ author: 
# @ created: 2018/7/21
 
import re
import sys
import os
 
reload(sys)
sys.setdefaultencoding("utf8")
 
 
def translate(str):
  out = set()
  line = str.strip().decode('utf-8', 'ignore') # 處理前進行相關的處理，包括轉(zhuǎn)換成Unicode等
  p2 = re.compile(ur'[^\u4e00-\u9fa5]') # 中文的編碼范圍是：\u4e00到\u9fa5
  zh = " ".join(p2.split(line)).strip()
  # zh = "\n".join(zh.split()) #dsds經(jīng)過相關處理后得到中文的文本
  for s in zh.split():
    out.add(s) # 經(jīng)過相關處理后得到中文的文本
  return out
 
def extract_file(path):
  result = set()
  try:
    f = open(path) # 打開文件
    lines = f.readlines()
    for line in lines:
      string = translate(line)
      if string:
        result.update(string)
  except Exception as e:
    pass
  return result
 
 
def extract(path):
  result = set()
  files = os.listdir(path)
  for file in files:
    if not file.startswith("."):
      if not os.path.isdir(path + "/" + file): # 判斷是否是文件夾，不是文件夾才打開ssgsg判斷是否是文件夾，不是文件夾才打開
        sub_file = extract_file(path + "/" + file)
        if sub_file:
          result.update(sub_file)
      else:
        print file
        child = extract(path + "/" + file)
        if child:
          result.update(child)
  return result
 
 
if __name__ == '__main__':
  path = "/Users/common"
  result = extract(path)
  res_file = open("result.txt", "w")
  for s in result:
    res_file.write(s + "\n")