cal_similarity2.py 文件源码

def extract(lines,filename,key_part=['# ??']):  # ???????????  ???????
    # print filename
    words=[]
    for key in key_part:
        try:
            # print key
            # print cchardet.detect(key)
            index=lines.index(key)
            print index
            words+=lines[index+1].decode('utf-8').split()  # ??????
        except Exception,e:
            print e
    words_dict={}
    for w in words:
        words_dict[w]=words_dict.get(w,0)+1   # dict
    # return words
    return Document(words_dict,filename)