Python script for removing files in folders

Question

I've been learning python for some time and I've come up with the idea to practice it a little. I created a script that removes files by their extension .html and .srt and also empty directories. It works good, but I've got a feeling that it could be improved. Can someone help me improve the code? Also I know it is not a good practice to use for-loop inside another for-loop. Files structure:

super secret folder
 |___empty folder
 |___folder1
 |___file01.jpg
 |___file02.srt
 |___file03.html
 |___folder2
 |___file01.jpg
 |___file02.srt
 |___file03.html
 |___folder3
 |___file01.jpg
 |___file02.srt
 |___file03.html

import os
mypath = "C:\\Users\\user\\super secret folder"
def extract_ext(file_path):
 split_tuple = os.path.splitext(file_path)
 return split_tuple[0], split_tuple[1]
def abs_path(my_file_path, file_name):
 return f"{my_file_path}\\{file_name}"
subfolders_list = os.listdir(mypath)
subfolders_paths = [os.path.join(mypath, path) for path in subfolders_list]
 
for folder_path in subfolders_paths:
 files_list = os.listdir(folder_path)
 
 if len(files_list) != 0: 
 
 for file_name in files_list:
 file_abs_path = abs_path(folder_path, file_name)
 _, file_extension = extract_ext(file_abs_path)
 
 
 if file_extension == ".html" or file_extension == ".srt":
 os.remove(file_abs_path)
 else:
 os.rmdir(folder_path)

• Reinderien: I think I'll untag "Windows", because: good news, this script can easily be made portable.
• lukstru: Python pathlib.Path overrides the division operator and adds OS specific separators: path1 / path2 = "path1Content\path2Content" on Windows and "path1Content/path2Content" on POSIX

Answer 1

Your code can be simplified through the usage of os.walk() and pathlib.Path.

from os import walk
from pathlib import Path
def cleanup(directory: Path, suffixes: set[str]) -> None:
 """Remove unwanted files with the given
 suffixes as well as empty directories.
 """
 
 for root, dirs, files in walk(directory):
 root = Path(root)
 for file in map(lambda filename: root / filename, files):
 if file.suffix in suffixes:
 file.unlink()
 if root.is_dir() and not dirs and not files:
 root.rmdir()
def main() -> None:
 cleanup(Path('folder'), {'.html', '.srt'})
if __name__ == '__main__':
 main()

This way it is more naturally readable, what the code does plus you have directory recursion. By using a separate function (cleanup()), you can also re-use the code with other directories or suffix sets.

Before

$ tree folder/
folder/
├── bar
│  ├── file01.jpg
│  ├── file02.html
│  └── file03.srt
├── foo
│  ├── file01.jpg
│  ├── file02.html
│  └── file03.srt
└── spamm

After

$ tree folder/
folder/
├── bar
│  └── file01.jpg
└── foo
 └── file01.jpg
2 directories, 2 files

Answer 2

For loops inside for loops are necessary for this use case, you can however decrease indention levels by extracting code blocks into methods:

for folder_path in subfolders_paths: 
 if not is_empty(folder_path): 
 remove_unwanted_files(folder_path)
 else:
 remove_path(folder_path)

Indention can further be decreased by extracting the conditional as well:

for folder_path in subfolders_paths:
 remove_empty_and_unwanted(folder_path)