dustr 0.1.11

Rust-based disk usage analyzer with histogram visualization

Dustr

Dustr is a Rust-based implementation of duk, a commandline utility that shows disk usage statistics in a directory with a histogram visualization.

Introduction

Dustr provides the same functionality as duk but with a Rust backend for improved performance. It will show you a histogram of the disk usage in a directory:

Statistics of directory "." :

Size           In %   Histogram            Name
4.0 KB         1.39  #                    .gitignore
4.0 KB         1.39  #                    setup.py
4.0 KB         1.39  #                    README.md
12.0 KB        4.17  ##                   dist/
12.0 KB        4.17  ##                   dustr/
12.0 KB        4.17  ##                   build/
16.0 KB        5.56  ##                   duk.egg-info/
220.0 KB       76.39 ####################  .git/

Total directory size: 284.0 KB

Installation

From PyPI

pip install dustr

From source

First, ensure you have Rust installed (see rustup.rs).

Then install maturin:

pip install maturin

Build and install the package:

cd dustr
maturin develop  # For development
# OR
maturin build --release  # To build a wheel
pip install target/wheels/dustr-*.whl

Usage

dustr [OPTIONS] [DIRECTORY]

Options:

• -i, --inodes: Show inode count instead of size
• -g, --nogrouping: Don't use thousand separators (for inode mode)
• -f, --noF: Don't add file type indicators (/ for directories, @ for symlinks)
• -j, --json: Output results as JSON

JSON output

dustr --json .

{
  "directory": ".",
  "mode": "size",
  "entries": [
    { "name": ".gitignore", "value": 4, "percentage": 1.41 },
    { "name": ".git/", "value": 220, "percentage": 77.46 }
  ],
  "total": 284
}

Differences from duk

• Performance: Rust backend with parallel directory traversal (jwalk + rayon)
• Implementation: Uses native Rust file system operations instead of calling du command
• Progress: Since the Rust implementation calculates all sizes in parallel, progress indication is less granular

Requirements

• Python >= 3.9
• Rust toolchain (for building from source)

License

LGPL-3.0-or-later