hownet_handel.py 文件源码

def jieba_cut():
    #??pos_all_dict??
    fp_pos = open("hownet/pos_all_dict.txt", "r")   # ?????????
    fp_pos_cut = codecs.open('hownet/pos_all_cut.txt', "w+", encoding='UTF-8')  # ????????????
    contents = fp_pos.readlines()
    for content in contents:
        word = content.decode("utf-8")  # ??
        word_tag = pseg.cut(word)
        str_tag = ""
        for tag in word_tag:
            str_tag += str(tag.word) + '/' + str(tag.flag)
        p = re.compile(r'/x(.*)')
        str_tag = p.sub(r'\1', str_tag)   # ??????
        fp_pos_cut.write(str_tag)
    fp_pos.close()
    fp_pos_cut.close()

    #??pos_all_dict??
    fp_neg = open("hownet/neg_all_dict.txt", "r")   # ?????????
    fp_neg_cut = codecs.open('hownet/neg_all_cut.txt', "w+", encoding='UTF-8')  # ????????????
    contents = fp_neg.readlines()
    for content in contents:
        word = content.decode("utf-8")  # ??
        word_tag = pseg.cut(word)
        str_tag = ""
        for tag in word_tag:
            str_tag += str(tag.word) + '/' + str(tag.flag)
        p = re.compile(r'/x(.*)')
        str_tag = p.sub(r'\1', str_tag)  # ??????
        fp_neg_cut.write(str_tag)
    fp_neg.close()
    fp_neg_cut.close()

# ????????????