Python struct模块读取表格数据bin文件

缺乏、安全感 2022-11-16 01:43 244阅读 0赞

bin文件是表格数据 ,数据类型为double, 每行有21个数据  
方法一: 逐个字节去读取, 写成csv后, 再读成pandas.Dataframe

import struct
    import os
    import pandas as pd
    
    
    def bin2csv(bin_file, col_num=21):
        """ bin文件转csv :param bin_file: bin文件路径 :param col_num: 字段数量, 默认21 :return: csv文件路径 """
        bfile, _ = os.path.splitext(bin_file)
        csv_file = f'{bfile}.csv'
        with open(bin_file, 'rb') as f:
            with open(csv_file, 'w') as g:
                n = 0
                while True:
                	# 每次读8字节
                    c = f.read(8)
                    if len(c) == 0:
                        break
                    # 解密bin数据
                    data,  = struct.unpack('d', c)
                    # 保留8位小数
                    data = round(data, 8)
                    n += 1
                    if n % col_num != 0:
                        g.write(str(data) + ',')
                    else:
                        g.write(str(data) + '\n')
        return csv_file
    
    
    if __name__ == '__main__':
        file = '~/Downloads/ains_imu2.bin'
        csv_file = bin2csv(file)
        col_list = [f'A{i}' for i in range(21)]
        df = pd.read_csv(csv_file, encoding='utf8', names=col_list, dtype=str)

方法二: 直接每次读取168个字节, 把21个数直接append到列表, 再将列表转成pandas.Dataframe

import pandas as pd
    import struct
    
    
    def bin2df(bin_file, col_num=21):
        dlist = []
        with open(bin_file, 'rb') as f:
            while True:
                try:
                    c = f.read(8 * col_num)
                    data = struct.unpack(col_num * 'd', c)
                    dlist.append(data)
                except struct.error as e:
                    print('读取完毕')
                    break
    
        df = pd.DataFrame(dlist, columns=[f'A{i}' for i in range(col_num)], dtype=str)
        return df
    
    
    if __name__ == '__main__':
        file = '~/Downloads/ains_imu2.bin'
        df = bin2df(file)

方法三: 直接使用numpy 读取成ndarray

import numpy as np
    data = np.fromfile(bin_file).reshape(-1, 21)