欢迎学习交流！
邮箱： z…@1…6.com
网站： https://zephyrhours.github.io/

一、问题描述

有时候为了方便，我们在进行程序编写时，会使用对话框来批量打开文件，但是使用os.listdir读取文件时，会出现文件乱序问题，此时直接使用sorted命令可能无法解决问题，尤其是遇到一些文件名称编码复杂时。具体如下;

• 下图是我们在window文件夹中打开时，文件排序方式；

• 下面是我们使用os.listdir命令获取的文件命名方式输出，具体代码和结果如下：

from tkinter import filedialog
import os

# 打开文件夹
folder_path = filedialog.askdirectory(title='Open the folder')
img_list = os.listdir(folder_path)
print('img_list:', img_list)

可以看到，此处的图像名称的排序方式是乱序的，这种情况下往往不方便我们后续的图像处理任务：

D:\anaconda3\envs\csdn\python.exe
C:\Users\zephy\Documents\Python\CSDN\demo_20240702.py
img_list:
[‘0_100000_A.jpg’,
‘0_100000_B.jpg’,
‘0_1000_A.jpg’,
‘0_1000_B.jpg’,
‘0_100_A.jpg’,
‘0_100_B.jpg’,
‘0_1100_A.jpg’,
‘0_1100_B.jpg’,
‘0_1200_A.jpg’,
‘0_1200_B.jpg’,
‘0_2000_A.jpg’,
‘0_2000_B.jpg’,
‘0_200_A.jpg’,
‘0_200_B.jpg’,
‘0_2100_A .jpg’,
‘0_2100_B.jpg’,
‘0_3000_A.jpg’,
‘0_3000_B.jpg’,
‘0_300_A.jpg’,
‘0_300_B.jpg’,
‘0_4000_A.jpg’,
‘0_4000_B.jpg’,
‘0_800_A.jpg’,
‘0_800_B.jpg’,
‘0_90000_A.jpg’,
‘0_90000_B.jpg’,
‘0_900_A.jpg’,
‘0_900_B.jpg’]

Process finished with exit code 0

二、os.listdir读取文件乱序问题解决方法

下面笔者提供一种解决方式，可以实现读取的文件顺序与文件夹浏览顺序保持一致。

from tkinter import filedialog
import os

# 打开文件夹
folder_path = filedialog.askdirectory(title='Open the folder')
img_list = os.listdir(folder_path)

# 以下命令中lambda表达式为解决问题关键
img_paths = sorted(img_list, key=lambda x: (len(x), x), reverse=False)

print('img_paths:', img_paths)

此次的输出结果如下：

D:\anaconda3\envs\csdn\python.exe C:\Users\zephy\Documents\Python\CSDN\demo_20240702.py
img_paths:
[‘0_100_A.jpg’,
‘0_100_B.jpg’,
‘0_200_A.jpg’,
‘0_200_B.jpg’,
‘0_300_A.jpg’,
‘0_300_B.jpg’,
‘0_800_A.jpg’,
‘0_800_B.jpg’,
‘0_900_A.jpg’,
‘0_900_B.jpg’,
‘0_1000_A.jpg’,
‘0_1000_B.jpg’,
‘0_1100_A.jpg’,
‘0_1100_B.jpg’,
‘0_1200_A.jpg’,
‘0_1200_B.jpg’,
‘0_2000_A.jpg’,
‘0_2000_B.jpg’,
‘0_2100_B.jpg’,
‘0_3000_A.jpg’,
‘0_3000_B.jpg’,
‘0_4000_A.jpg’,
‘0_4000_B.jpg’,
‘0_2100_A .jpg’,
‘0_90000_A.jpg’,
‘0_90000_B.jpg’,
‘0_100000_A.jpg’,
‘0_100000_B.jpg’]
Process finished with exit code 0

此时可以看到，输出的文件名称与我们所期望的（window文件夹中打开时默认的排序）保持了一致。

下面还可以通过设置，实现需求数据的筛选，详细代码如下：

from tkinter import filedialog
import os

# 打开文件夹
folder_path = filedialog.askdirectory(title='Open the folder')
img_list = os.listdir(folder_path)
# print('img_list:', img_list)

filter_str = '_A.jpg'

img_names = []
img_paths = sorted(img_list, key=lambda x: (len(os.path.basename(x)),os.path.basename(x)), reverse=False)
# print('img_paths:', img_paths)

for i in range(len(img_paths)):
    if img_paths[i].find(filter_str) != -1:
        img_names.append(img_paths[i])

img_names = [os.path.join(folder_path, img_names[i]) for i in range(len(img_names))]
print('img_names:', img_names)

具体结果如下：

D:\anaconda3\envs\csdn\python.exe
C:\Users\zephy\Documents\Python\CSDN\demo_20240702.py img_names: [
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_100_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_200_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_300_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_800_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_900_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_1000_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_1100_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_1200_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_2000_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_3000_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_4000_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_90000_A.jpg’,
‘C:/Users/zephy/Desktop/CSDN/datasets/test\0_100000_A.jpg’] Process
finished with exit code 0