kompress 0.3.20260707

kompress lets you transparently decompress common archive formats while using pathlib.Path objects.

It attempts to keep the API as close to pathlib, but things may not be perfect yet

The common case for this is:

• You have a compressed file that when decompressed contains some text
• You have a function (perhaps from another library, that you don't control), that knows how to take a pathlib.Path object and call .read() or .read_text() on it

Without wrapping the function or adding logic to let it read from compressed files, this lets you do the following:

from pathlib import Path
from kompress import CPath, is_compressed

def char_count(path: Path) -> int:
    # here, if a CPath is passed, it decompresses and returns a string
    return len(path.read_text())

inp = input("Enter a file to process: ")  # file came from user
char_count(CPath(inp))

This currently supports these archive formats:

• .xz
• .zip
• .zst
• .tar.gz
• .gz
• .zstd (since python3.14, on older version use pip install 'kompress[zstd]')
• .lz4 (pip install 'kompress[lz4]')

If it doesn't recognize the filetype, it will just call pathlib.Path.open, reading it like a regular file.

Originally discussed in this HPI issue

Installing

pip install kompress