Two APIs: the modern object-oriented pathlib (recommended) and the older string-based os / os.path. Both work everywhere; pathlib is easier to read and handles / vs \ for you.
See also: files (reading/writing files) · loops
from pathlib import Path
p = Path('C:/Windows/System32') / 'calc.exe' # join with /
p # -> WindowsPath('C:/Windows/System32/calc.exe')
str(p) # -> 'C:\\Windows\\System32\\calc.exe'
p.name # -> 'calc.exe'
p.stem # -> 'calc'
p.suffix # -> '.exe'
p.parent # -> WindowsPath('C:/Windows/System32')
p.parents[1] # -> WindowsPath('C:/Windows')
p.parts # -> ('C:\\', 'Windows', 'System32', 'calc.exe')
p.anchor # -> 'C:\\'
p.with_suffix('.bak') # -> WindowsPath('C:/Windows/System32/calc.bak')
p.with_name('notepad.exe') # -> WindowsPath('C:/Windows/System32/notepad.exe')
Path.cwd() # current working directory
Path.home() # -> WindowsPath('C:/Users/me')
Path('~/docs').expanduser() # -> WindowsPath('C:/Users/me/docs')
Path(__file__).resolve().parent # folder of the running script (common idiom)On Linux/macOS you get PosixPath instead of WindowsPath; the API is the same.
os / os.path |
pathlib |
|---|---|
os.path.join(a, b) |
Path(a) / b |
os.getcwd() |
Path.cwd() |
os.path.abspath(p) |
Path(p).resolve() (or .absolute()) |
os.path.basename(p) |
p.name |
os.path.dirname(p) |
p.parent |
os.path.splitext(p) |
p.stem, p.suffix |
os.path.exists(p) |
p.exists() |
os.path.isfile(p) / isdir(p) |
p.is_file() / p.is_dir() |
os.path.getsize(p) |
p.stat().st_size |
os.makedirs(p, exist_ok=True) |
p.mkdir(parents=True, exist_ok=True) |
os.listdir(p) |
p.iterdir() |
glob.glob('**/*.txt', recursive=True) |
p.rglob('*.txt') / p.glob('**/*.txt') |
os.walk(p) |
p.walk() (3.12+) |
os.remove(p) / os.rmdir(p) |
p.unlink() / p.rmdir() |
os.rename(a, b) |
Path(a).rename(b) |
os.path.relpath(p, start) |
p.relative_to(start) (walk_up=True 3.12+) |
import os
os.path.join('usr', 'bin', 'spam')
# 'usr\\bin\\spam' (Windows)
# 'usr/bin/spam' (Linux/macOS)my_files = ['accounts.txt', 'details.csv', 'invite.docx']
for filename in my_files:
print(os.path.join('C:\\Users\\asweigart', filename))
# C:\Users\asweigart\accounts.txt
# C:\Users\asweigart\details.csv
# C:\Users\asweigart\invite.docxSame with pathlib:
from pathlib import Path
base = Path('C:/Users/asweigart')
[base / f for f in my_files]
# -> [WindowsPath('C:/Users/asweigart/accounts.txt'), WindowsPath('C:/Users/asweigart/details.csv'), ...]💡 In Windows paths use raw strings (
r'C:\Users\me') or forward slashes ('C:/Users/me') to avoid escaping every backslash.
💡 There are also the dot (
.) and dot-dot (..) folders. These are not real folders but special names that can be used in a path. A single period ("dot") for a folder name is shorthand for "this directory." Two periods ("dot-dot") means "the parent folder."
os.path.abspath('.')
# 'C:\\Python313'
os.path.abspath('.\\Scripts')
# 'C:\\Python313\\Scripts'
os.path.isabs('.')
# False
os.path.isabs(os.path.abspath('.'))
# True
Path('.').resolve() # -> WindowsPath('C:/Python313') (absolute, resolves .. and symlinks)
Path('.').is_absolute() # -> Falseos.path.relpath('C:\\Windows', 'C:\\')
# 'Windows'
os.path.relpath('C:\\Windows', 'C:\\spam\\eggs')
# '..\\..\\Windows'
Path('C:/Windows/System32').relative_to('C:/Windows')
# -> WindowsPath('System32')
Path('C:/Windows').relative_to('C:/spam/eggs', walk_up=True) # 3.12+
# -> WindowsPath('../../Windows')import os
os.getcwd()
# 'C:\\Python313'
Path.cwd()
# -> WindowsPath('C:/Python313')os.chdir('C:\\Windows\\System32')
os.getcwd()
# 'C:\\Windows\\System32'💡 Changing the cwd affects the whole process. Prefer building absolute paths (e.g.
Path(__file__).parent / 'data.txt') overchdir.
import os
os.makedirs('C:\\delicious\\walnut\\waffles', exist_ok=True) # creates intermediate folders
os.mkdir('C:\\delicious\\other') # only the last level
Path('C:/delicious/walnut/waffles').mkdir(parents=True, exist_ok=True)Without exist_ok=True they raise FileExistsError if the folder already exists.
import shutil
from pathlib import Path
Path('old.txt').unlink(missing_ok=True) # delete a file
Path('empty_dir').rmdir() # delete an EMPTY dir
shutil.rmtree('build') # ⚠️ delete a dir and everything inside
shutil.copy('a.txt', 'backup/') # copy file (into a folder or to a new name)
shutil.copy2('a.txt', 'b.txt') # also copies metadata (timestamps)
shutil.copytree('src', 'dst', dirs_exist_ok=True) # copy a whole tree
shutil.move('a.txt', 'archive/a.txt') # move / rename (across drives too)
Path('a.txt').rename('b.txt') # rename (same drive)path = 'C:\\Windows\\System32\\calc.exe'
os.path.basename(path)
# 'calc.exe'
os.path.dirname(path)
# 'C:\\Windows\\System32'
os.path.splitext(path)
# ('C:\\Windows\\System32\\calc', '.exe')If you need a path's dir name and base name together, you can just call os.path.split() to get a tuple value with these two strings, like so:
calc_file_path = 'C:\\Windows\\System32\\calc.exe'
os.path.split(calc_file_path)
# ('C:\\Windows\\System32', 'calc.exe')os.path.split() does not return a list of strings of each folder. For that, split on os.sep (or use Path(...).parts).
calc_file_path.split(os.sep)
# ['C:', 'Windows', 'System32', 'calc.exe']
Path(calc_file_path).parts
# ('C:\\', 'Windows', 'System32', 'calc.exe')On macOS and Linux systems, there will be a blank string at the start of the returned list:
'/usr/bin'.split(os.sep)
# ['', 'usr', 'bin']os.path.getsize('C:\\Windows\\System32\\calc.exe')
# 49152
Path('C:/Windows/System32/calc.exe').stat().st_size
# 49152os.path.exists('C:\\Windows')
# True
os.path.exists('C:\\some_made_up_folder')
# False
os.path.isdir('C:\\Windows\\System32')
# True
os.path.isfile('C:\\Windows\\System32')
# False
os.path.isdir('C:\\Windows\\System32\\calc.exe')
# False
os.path.isfile('C:\\Windows\\System32\\calc.exe')
# True
p = Path('C:/Windows/System32/calc.exe')
p.exists(), p.is_file(), p.is_dir()
# (True, True, False)It returns a list of all the files and subdirectories (names only, not full paths) in the given path, in arbitrary order.
os.listdir('.')
# ['.DS_Store', 'Playground', 'Projects', 'MDFiles', 'pruebas']
sorted(p.name for p in Path('.').iterdir()) # Path objects -> names
[p for p in Path('.').iterdir() if p.is_dir()] # only folders💡
os.scandir()is faster thanlistdir()+os.pathcalls when you need file type/size: each entry has.name,.path,.is_file(),.stat().
Example, total size of the files in a folder:
total_size = 0
for filename in os.listdir('C:\\Windows\\System32'):
total_size += os.path.getsize(os.path.join('C:\\Windows\\System32', filename))
print(total_size)
# 1117846456
# pathlib version (files only)
sum(f.stat().st_size for f in Path('C:/Windows/System32').iterdir() if f.is_file())os.walk(top, topdown=True, onerror=None, followlinks=False)Generate the file names in a directory tree by walking the tree either top-down or bottom-up. For each directory in the tree rooted at directory top (including top itself), it yields a 3-tuple (dirpath, dirnames, filenames).
Example #1:
Get the list of all files in directory tree at given path:
dir_name = 'c:\\projects\\hc2'
list_of_files = []
for dirpath, dirnames, filenames in os.walk(dir_name):
list_of_files += [os.path.join(dirpath, file) for file in filenames]Example #2:
List all .txt files in a specified directory + subdirectories:
import os
path = 'c:\\projects\\hc2\\'
files = []
# r=root, d=directories, f=files
for r, d, f in os.walk(path):
for file in f:
if file.endswith('.txt'):
files.append(os.path.join(r, file))
for f in files:
print(f)
# c:\projects\hc2\app\readme.txt
# c:\projects\hc2\whois\download\afrinic.txt
# c:\projects\hc2\whois\download\apnic.txt
# ...💡 With
topdown=Trueyou can prune the walk by modifyingdirnamesin place:dirnames[:] = [d for d in dirnames if d not in {'.git', '.venv', 'node_modules'}].
Path.walk() (3.12+) does the same but yields Path objects:
from pathlib import Path
for root, dirs, files in Path('c:/projects/hc2').walk():
dirs[:] = [d for d in dirs if not d.startswith('.')] # skip hidden folders
for name in files:
if name.endswith('.txt'):
print(root / name)glob.glob(pathname, *, recursive=False)The glob module finds all the pathnames matching a specified pattern according to the rules used by the Unix shell; results are returned in arbitrary order. No tilde expansion is done, but *, ?, and character ranges expressed with [] will be correctly matched.
For a literal match, wrap the meta-characters in brackets. For example, '[?]' matches the character '?'.
For example, consider a directory containing the following files: 1.gif, 2.txt, card.gif and a subdirectory sub which contains only the file 3.txt. glob() will produce the following results:
import glob
glob.glob('./[0-9].*')
# ['./1.gif', './2.txt']
glob.glob('*.gif')
# ['1.gif', 'card.gif']
glob.glob('?.gif')
# ['1.gif']
glob.glob('**/*.txt', recursive=True)
# ['2.txt', 'sub/3.txt']
glob.glob('./**/', recursive=True)
# ['./', './sub/'](On Windows the separators in the results are \: 'sub\\3.txt'.)
Example, list all .txt files in a directory + subdirectories (**):
import glob
path = 'c:\\projects\\hc2\\'
for f in glob.glob(path + '**/*.txt', recursive=True):
print(f)
# c:\projects\hc2\app\readme.txt
# c:\projects\hc2\whois\download\afrinic.txt
# ...And all directories (a pattern ending in a separator only matches folders):
for d in glob.glob(path + '**/', recursive=True):
print(d)
# c:\projects\hc2\
# c:\projects\hc2\analyzer\
# c:\projects\hc2\analyzer\out\
# ...The pathlib equivalents return Path objects (generators):
from pathlib import Path
root = Path('c:/projects/hc2')
list(root.glob('*.txt')) # this folder only
list(root.rglob('*.txt')) # recursive (= glob('**/*.txt'))
sorted(root.rglob('*.py'), key=lambda p: p.stat().st_size, reverse=True)[:5] # 5 biggest .py filesdef get_list_of_files(dir_name: str) -> list[str]:
"""Return all file paths below dir_name (recursive)."""
all_files = []
for entry in os.listdir(dir_name):
full_path = os.path.join(dir_name, entry)
# If entry is a directory then get the list of files in this directory
if os.path.isdir(full_path):
all_files += get_list_of_files(full_path)
else:
all_files.append(full_path)
return all_files
list_of_files = get_list_of_files('/home/varun/Downloads')import tempfile
from pathlib import Path
with tempfile.TemporaryDirectory() as tmp:
p = Path(tmp) / 'test.txt'
p.write_text('hi', encoding='utf-8')
print(p.exists()) # -> True
# the folder and its contents are deleted here