`pathlib` represents filesystem locations as `Path` objects. It provides an object-oriented alternative to `os` and `os.path` for building paths, inspecting files and directories, and performing filesystem operations. Available since Python 3.5, it often makes the same tasks more concise. section 12.1
Build and inspect a path
Use `from pathlib import Path` and `path = Path('some', 'directory', 'path.jpg')`. Then `path.name` is `'path.jpg'`, `path.parent` is `Path('some/directory')`, and `path.suffix` is `'.jpg'`. `path.parts` gives the components as a tuple. Build further paths with `path.joinpath(...)` or the `/` operator; the resulting path follows the operating system’s path conventions. section 12.2.5
Common operations stay attached to the path they act on:
- `Path.cwd()` gets the current directory. Creating another `Path` lets you work with a different location, but does not change the process’s current directory. section 12.2.3
- `iterdir()` yields paths in a directory; `glob(pattern)` yields matching paths. Use `rglob(pattern)` to search subdirectories. section 12.4.1 section 12.6.2
- `rename(new_path)` can move a file to another directory. `mkdir(parents=True)` creates needed intermediate directories, while `mkdir(exist_ok=True)` accepts an existing directory. section 12.4.1 section 12.6.2
- `unlink()` deletes a file, not a directory; `rmdir()` removes only an empty directory. section 12.4.1
Total matching file sizes
Starting with `root = Path('sample_data')`, loop over `root.rglob('*.test')`. For each path that is not a symlink (`not path.is_symlink()`), add `path.stat().st_size` to a running total. The search, symlink check and size lookup all use methods on path objects. section 12.6.2
Listing 20.1 File files_01.py
import datetime
import pathlib
FILE_PATTERN = "*.txt" #1
ARCHIVE = "archive"
def main():
date_string = datetime.date.today().strftime("%Y-%m-%d") #2
cur_path = pathlib.Path(".")
archive_path = cur_path.joinpath(ARCHIVE)
archive_path.mkdir(exist_ok=True) #3
paths = cur_path.glob(FILE_PATTERN)
for path in paths:
new_filename = f"{path.stem}_{date_string}{path.suffix}"
new_path = archive_path.joinpath(
new_filename) #4
path.rename(new_path) #5
if __name__ == '__main__':
main()
A complete archiving script finds matching text files, creates an archive directory, adds a date to each filename using `stem` and `suffix`, then moves each file with `rename()`. section 20.2
