Find all the Chinese characters in a file mixed with Chinese, Japanese and English using langid

def find_chinese(filename):
    text = open(filename, "r").read()
    text = re.sub("[\s+\.\!\/_,$%^*(+\"\'“”《》?“]+|[+——！，。？、~@#￥%……&*（）]+", "",text)
    result = ""
    for i in text:
        if langid.classify(i)[0]=="zh":
            result += i
    return result