# -*- coding: utf-8 -*-
"""
Created on Wed Mar 28 17:06:34 2018

@author: Administrator
"""
import sys
import thulac
import sys
import os
import io
thu=thulac.thulac(seg_only=True)
for i in range(0, 19):
    path2="E:/社交网络/weiboAPIsprider/数据集/"#原数据集
    f = io.open(path2+str(i)+'.txt', "r" , encoding='UTF-8')
    text = ""
    for line in f:
        text = text+line
    print(text)
    seg_list = thu.cut(text, Text=False)#处理过的数据集
    with io.open("E:/社交网络/jiebaTest/原始结果集/"+'thulac'+str(i)+'.txt', 'w', encoding='UTF-8') as file:
     file.write('\n'.join(seg_list))#结果写入文件
     file.close()