Chonk

A desktop app (Rust + gpui) that reads the measurements of a Tanita body composition scale (tested with the BC-601) from its SD card, archives them in a local SQLite database and shows them as charts. Runs on macOS, Linux and Windows. The interface is available in English and German.

The screenshot shows made-up sample data.

Install

Ready-made packages are attached to the releases:

System	Package	Note
macOS	Chonk-…-macos-universal.zip	The app is not notarized. On first launch, right-click and choose "Open", or run xattr -dr com.apple.quarantine Chonk.app.
Linux	chonk-…-linux-x86_64.tar.gz	Needs Vulkan and either Wayland or X11.
Windows	chonk-…-windows-x86_64.zip	Run chonk.exe directly.

On Arch Linux, build and install the package from the PKGBUILD with paru:

git clone https://github.com/sebhofmann/chonk.git
cd chonk/pkg/arch
paru -Bi .

Build it yourself

cargo run --release

On Debian/Ubuntu, run script/linux-deps.sh first.

Usage

Chonk looks for the folder TANITA/GRAPHV1 on all removable drives (macOS: /Volumes, Linux: /run/media/$USER, /media, /mnt, Windows: every drive letter). When the card is present, at startup or later, new measurements are imported automatically. Measurements that are already archived are skipped, and the card is only ever read.

On first launch the language follows your system. You can switch it in the bottom right corner.

The scale knows eight profiles but no names. Click the profile name below the title to give it a name of your own (Enter saves, Esc cancels).

Data

What	Where
Archive	~/.chonk/chonk.db (SQLite), on Windows %USERPROFILE%\.chonk
Measurements	Table measurements, unique per profile and timestamp, including the original line in raw
Profiles	Table profiles (slots 1–8 of the scale, custom name in name)
Settings	Table settings (for example the language)
Import log	Table imports

The app used to be called "Fettsack". An existing archive in ~/.fettsack is moved to ~/.chonk automatically on first launch.

Environment variables for testing: CHONK_HOME (a different data folder) and CHONK_CARD (path to a TANITA/GRAPHV1 folder instead of the card).

script/fake-card.py creates a card with made-up measurements for two profiles:

python3 script/fake-card.py /tmp/demo/TANITA/GRAPHV1
CHONK_HOME=/tmp/demo/home CHONK_CARD=/tmp/demo/TANITA/GRAPHV1 cargo run

The scale's file format

DATA/DATAn.CSV holds one line of key-value pairs per measurement, SYSTEM/PROFn.CSV holds the profile. n is the profile slot.

Key	Meaning
MO	Model
DT, Ti	Date (DD/MM/YYYY) and time
Bt, GE, AG, Hm, AL	Body type, gender, age, height in cm, activity level
Wk, MI	Weight in kg, BMI
FW	Total body fat in %
Fr, Fl, FR, FL, FT	Fat in % for right/left arm, right/left leg, trunk
mW	Total muscle mass in kg
mr, ml, mR, mL, mT	Muscle mass in kg, same segments
bW	Bone mass in kg
IF	Visceral fat level
rD, rA	Daily calorie need in kcal, metabolic age
ww	Body water in %
CS	Checksum: sum of all bytes before CS modulo 256, hexadecimal

Lines with a wrong checksum are skipped during import.

Development

• gpui is built with the runtime_shaders feature. On macOS the Xcode Command Line Tools are therefore enough, a full Xcode install is not needed.
• Translations live in src/i18n.rs. Fixed texts are fields of a struct, so the compiler reports missing translations.
• The CI workflow checks every push on Linux, macOS and Windows (formatting, Clippy, tests). Dependencies are cached between runs, and documentation-only changes do not trigger a run.
• The Release workflow starts when a release is published on GitHub (Releases → "Draft a new release" → pick a tag such as v0.1.0 → "Publish release"). It builds the packages and attaches them to the release.

License

MIT