# -*- coding: UTF-8 -*-
import jieba.posseg
import tensorflow as tf
import pandas as pd
import csv
import math
"""
1.必須獲取CSV資料夾（ID：文字）
2.返回(ID：分詞後的文字)
"""
flags = tf.app.flags
flags.DEFINE_string("train_file_address","D:/NLPWORD/cut_word_test/hzytest.csv","新增訓練資料檔案")
flags.DEFINE_string("result_file_address"
 
,"D:/NLPWORD/cut_word_test/hzytest_result.csv","生成結果資料檔案")
FLAGS = tf.app.flags.FLAGS
def cut_word(train_data):
    """
        把資料按照行進行遍歷，然後把結果按照行寫在csv中
        :return:分詞結果list
        """
    jieba.load_userdict("newdict.txt")
    with open(FLAGS.result_file_address, "w", encoding='utf8') as csvfile:
        writer = csv.writer(csvfile)
        for
 
 row in train_data.index:
            datas = train_data.loc[row].values[1]
            if  isinstance(datas,str) or not math.isnan(datas):
                words = jieba.posseg.cut(datas)
                line = ''
                for word in words:
                    line = line + word.word + " "
                writer.writerow([train_data.loc[row].values[0
 
], line])
def main(_):
    data = pd.read_csv(FLAGS.train_file_address)
    cut_word(data)

if __name__ == "__main__":
    tf.app.run(main)