kmeans_cluster.py 文件源码

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