Working With Absolute and Relative Paths

Denis Zastanceanu, CC BY-SA 4.0, via Wikimedia Commons

Shell #

Getting the current working directory #

The environment variable $PWD and the command pwd are essentially synonymous. pwd has flags for dealing with symlinks.

Absolute (canonical) path #

• realpath - Turns relative paths into absolute paths
• readlink -f - Originally meant to read symlinks. The -f flag means give “cannonical” path.

Elements of a directory #

• dirname - Strips the rightmost part from a given directory.
• basename - Strips all base directories, leaving the last (rightmost) part of a path/file

NOTE: These commands do not check if a file/directory actually exists, they just manipulate path strings. Also, be careful when dealing with symbolic links and read the man page to get the desired behavior.

Python #

Using the os module #

import os

Current working directory #

os.getcwd()

Absolute (cannonical) path #

os.path.abspath("blog")

Elements of a directory #

os.path.dirname("blog")
os.path.basename("blog")

Pathlib #

As of Python 3.4 here is a module called pathlib, which provides a more OOP style of interacting with files.

os and pathlib have lots of overlap but the module provides extra methods for file globbing, pattern matching does a better job of integrating “system paths” into the language IMO.

This module converts paths from strings into higher level “file objects”. This is something that the os module does not do – and perhaps you don’t need this extra encapsulation.

If you ever find yourself importing re or shutil to deal with files and paths in Python, it might be worth it to try out pathlib.

I’ll give some example usage:

from pathlib import Path

# get (absolute) current directory
wd = Path.cwd()

# the resolve method converts any
# relative into an absolute path
wd = Path('.').resolve()

# adding subdirectories
# instead of using os.path.join()
# you can simply use the / operator
n = wd / 'content'

# print basename and dirname
print(wd.parent, wd.name)

# recursively globbing filenames
# from a relative path
for f in wd.glob('**/*.html'):
  print(f)

list(wd.rglob('*.html'))