import re
import os


# 按名称排序，获取目录下所有文件
def get_file_list(file_path):
    dir_list = os.listdir(file_path)
    if not dir_list:
        return
    else:
        # 注意，这里使用lambda表达式，key设为下列数值可按照修改时间排序
        # os.path.getmtime() 函数是获取文件最后修改时间
        # os.path.getctime() 函数是获取文件最后创建时间
        dir_list = sorted(dir_list, key=lambda x: int(x[:-4]))  # 去掉.txt
        return dir_list


def killAnUnseen(s):
    try:
        s.encode('gbk')
        # print("return" + s)
        return s
    except UnicodeEncodeError as err:
        mode = re.findall(r'position ([0-9]*): illegal multibyte sequence', str(err))
        # if mode:
        #     print("position:" + mode[0])
        # print(s[int(mode[0])])
        news = s.replace(s[int(mode[0])], "", 1)
        return killAnUnseen(news)


filelist = get_file_list('./data_old')
# filelist = filelist[60:]
for file in filelist:
    print("\r {} is working".format(file), end="")
    result = []
    with open('./data_old/' + file, 'r') as f:  # py3默认utf-8编码
        lines = f.readlines()
        for line in lines:
            try:  # 尝试gbk编码
                line.encode('gbk')
            except UnicodeEncodeError:  # 编码失败，去掉无法编码的字符
                print(line)
                line = killAnUnseen(line)
            result.append(line)
    with open('./data/' + file, 'w', encoding='gbk') as f_new:
        f_new.writelines(result)

def killAnUnseen(s):
    try:
        s.encode('gbk')
        # print("return" + s)
        return s
    except UnicodeEncodeError as err:
        mode = re.findall(r'position ([0-9]*): illegal multibyte sequence', str(err))
        # if mode:
        #     print("position:" + mode[0])
        # print(s[int(mode[0])])
        news = s.replace(s[int(mode[0])], "", 1)
        killAnUnseen(news)  # 这里应该写成return killAnUnseen(news)