import os

The os module in Python provides functions to interact with the operating system (OS).
It allows you to work with file systems, environment variables, process management, and system commands.

Path and File Operations

os.getcwd()

Linux uses slashes (/) as path separators.
Windows uses backslashes (\) as path separators. Since Python understands forward slashes (/) even on Windows, use slashes(/) for os.path related operations.

OS	os.getcwd()	os.getcwd().replace(“\”, “/”)
Windows	C:\Users\Username\projects	/home/username/projects
Linux	/home/username/projects	/home/username/projects

import os

# Linux
path = os.getcwd()
print(path) # /home/username/projects

# Windows
path = os.getcwd()
print(path) # C:\Users\Username\projects
print(path.replace("\\", "/")) # C:/Users/Username/projects

os.path.exists()

Check if a File or Directory Exists.

import os

# Linux
print(os.path.exists("/home/user/projects/data.txt"))  # Returns True or False

# Windows
# As I told you, Python understands forward slashes (`/`) even on Windows
print(os.path.exists("C:\Users\Username\projects/data.txt"))  # Returns True or False

os.path.isfile() and os.path.isdir()

Check if it’s a File or a Directory.

import os

print(os.path.isfile("/home/user/projects/data.txt"))  # True if it's a file
print(os.path.isdir("/home/user/projects/data"))       # True if it's a directory

os.listdir()

List Files in a Directory.

import os

print(os.listdir("/home/user/projects"))  # Returns a list of files and folders

['.dockerignore', '.env', '.flake8', '.gitignore', 'commons', 'config', 'dockerfile', 'dockertest_main.py', 'log', 'pretrained_models', 'readme.md', 'requirements.txt', 'tasks', 'zMinioTest.txt', 'zMyTest.ipynb']

os.makedirs()

Create a Directory. For reference, os.mkdir is also available, but it cannot create nested directories. If possible, use os.makedirs instead.
The exist_ok=True argument is used in functions like os.makedirs() to prevent errors when a directory already exists.

import os

# Creates all parent directories if they don’t exist. The `exist_ok=True` argument is used in functions like `os.makedirs()` to prevent errors when a directory already exists.
os.makedirs("parent/child", exist_ok=True)  
# FileNotFoundError: [WinError 3] 지정된 경로를 찾을 수 없습니다
os.mkdir("parent/child")  

os.remove()

Delete a File.

import os

os.remove("file.txt")

os.rmdir()

Delete an Empty Directory.

import os

os.rmdir("empty_folder")  # Deletes the empty folder
# An error occurs if the folder is not empty.
# OSError: [WinError 145] 디렉터리가 비어 있지 않습니다: 'empty_folder'

os.environ.get()

Get an Environment Variable.

import os

# e.g. home environment
print(os.environ.get("HOME"))  # Linux/macOS, /home/jaoneol
print(os.environ.get("USERPROFILE"))  # Windows, C:\Users\ycjang

Path Handling Module

os.path.dirname()

Returns the directory path containing target(file or directory) (i.e., the full path of the parent directory).

import os

# e.g. target file
remotefile = "/home/user/projects/data/file.txt"
print(f'remotefile:{remotefile}') # /home/user/projects/data/file.txt  

path = os.path.dirname(remotefile) 
print(f'path:{path}') # /home/user/projects/data 

parts = path.split('/')
print(f'parts:{parts}') # ['', 'home', 'user', 'projects', 'data']

# e.g. target directory
remotefile = "/home/user/projects/data/mydata"
print(f'remotefile:{remotefile}') # /home/user/projects/data/mydata

path = os.path.dirname(remotefile) 
print(f'path:{path}') # /home/user/projects/data

parts = path.split('/')
print(f'parts:{parts}') # ['', 'home', 'user', 'projects', 'data']

os.path.basename()

Returns only the filename.

import os

# e.g. target file
remotefile = "/home/user/projects/data/file.txt"

onlyfilename = os.path.basename(remotefile) # /home/user/projects/data
print(f'onlyfilename:{onlyfilename}') # file.txt

remotefile = "/home/user/projects/data/file.tar.gz" 

onlyfilename = os.path.basename(remotefile) # /home/user/projects/data
print(f'onlyfilename:{onlyfilename}') # file.tar.gz

os.path.join()

os.path.join() automatically handles OS-specific path separators (/ or \), making it cross-platform compatible.

import os

# Linux
path = os.path.join("/home/user/projects", "data", "file.txt")
print(path) # /home/user/projects/data/file.txt

# Windows
path = os.path.join("/home/user/projects", "data", "file.txt")
print(path) # /home/user/projects\data\file.txt
path = path.replace("\\", "/")
print(path) # /home/user/projects\data\file.txt

os.path.splitext()

Split File Extension.

import os

filename = "/home/user/projects/data/file.txt"
name, ext = os.path.splitext(filename)
print(name) # "/home/user/projects/data/file"
print(ext)  # ".txt"

filename = "/home/user/projects/data/file.tar.gz"
name, ext = os.path.splitext(filename)
print(name)  # "/home/user/projects/data/file.tar"
print(ext)   # ".gz"

os.path.split()

Splits the path into (directory, filename).

import os

filename = "/home/user/projects/data/file.txt"
directory, file_name  = os.path.split(filename)
print(directory) # "/home/user/projects/data"
print(file_name) # "file.txt"

# If there is no trailing slash (`/`), the last element is returned as the "filename."
filename = "/home/user/projects/data/mydata"
directory, file_name = os.path.split(filename)
print(directory) # "/home/user/projects/data"
print(file_name) # "mydata"

# If there is a trailing slash (`/`), the last element is recognized as a directory, and an empty string (`""`) is returned.
filename = "/home/user/projects/data/mydata/"
directory, file_name = os.path.split(filename)
print(directory) # "/home/user/projects/data/mydata"
print(file_name) # ""

Str Handling Module

str.split()

string = "/home/user/projects/data/file.tar.gz"
result = string.split() # Split based on spaces.
print(result) # ['/home/user/projects/data/file.tar.gz']  

result = string.split(".")
print(result) # ['/home/user/projects/data/file', 'tar', 'gz']

result = string.split("/")
print(result) # ['', 'home', 'user', 'projects', 'data', 'file.tar.gz']

str.removeprefix() and str.removesuffix()

string = "/home/user/projects/data/file.tar.gz"
result = string.removeprefix("/home")
print(result) # /user/projects/data/file.tar.gz  

result = string.removeprefix("home")
print(result) # /home/user/projects/data/file.tar.gz  

result = string.removesuffix(".tar.gz")
print(result) # /user/projects/data/file