Dev Blog

Paths and text files ​

Paths and pathlib file operations ​

Path is a path object; the / operator joins a component without manually choosing a Windows or Unix separator. Path.cwd() returns the process's working directory, and Path.home() returns the user's home directory. They do not necessarily match the program file's location. For a resource next to a module, you can use Path(__file__).resolve().parent / "data.json".

A relative path is interpreted from the working directory. This is why a program may work in an IDE but fail to find a file in a terminal. Accept the path to a user file as a command argument. For a small test, pass a Path into the function; do not hide an absolute path from your own computer inside the function body.

mkdir(parents=True, exist_ok=True) creates a directory and its parents. iterdir() iterates over immediate children, glob("*.csv") over matching names, and rglob("*.csv") recursively. Do not assume traversal order is sorted: use sorted for a reproducible report. stem returns the name without its last suffix, suffix returns the suffix, and stat().st_size returns the file size in bytes.

exists() or is_file() does not guarantee that the next read will succeed: the state can change between operations. Handle OSError and its specific subclasses where you perform the action. unlink() deletes a file, so run such operations only in a separate test folder during labs. Documentation: https://docs.python.org/3.14/library/pathlib.html.

Text files and encoding ​

open returns a file object. Mode r reads, w creates or truncates an existing file, a appends, and x creates a new file only. Adding b means binary data. Specify encoding="utf-8" for a text file; binary mode works with bytes and does not accept a text encoding. Use an open resource in with.

read() reads the entire file and is suitable for small datasets. A for line in stream loop processes a file line by line. A line may contain a trailing newline character, so use strip deliberately: it also removes leading and trailing spaces, which are sometimes part of the data. Path methods read_text and write_text are concise forms of reading and writing the entire contents.

py
from pathlib import Path
from tempfile import TemporaryDirectory


with TemporaryDirectory() as folder:
    path = Path(folder) / "notes.txt"
    path.write_text("First\nSecond\n", encoding="utf-8")
    with path.open(encoding="utf-8") as stream:
        for number, line in enumerate(stream, 1):
            print(number, line.rstrip("\n"))
1 First
2 Second

The temporary directory here isolates the demonstration from real files. For persistent storage, use a path supplied by the user. FileNotFoundError means the file is absent, PermissionError means access is denied, and IsADirectoryError means an attempt to treat a directory as a file on platforms that report it this way. UnicodeError may indicate a different encoding. Do not replace all these causes with an empty list: otherwise, a corrupted file will look like an empty database.

datetime works with dates and time intervals, os with the process environment and system interfaces, and shutil with copying files and trees. Choose pathlib for paths and shutil.copy2 for copying when available metadata is needed. No utility library removes the need to check exactly which directory will be changed.