Basquiat

Small Mojo 🔥 and Python library for creative coding powered by Cairo (through pycairo). The name is a humble homage to the artist Jean-Michel Basquiat.

Both Mojo and Python implementations depend on pycairo. I'm planning to migrate to cffi libcairo directly from Mojo when possible, but we already have the Mojo implementation running almost 3x faster than Python:

% time mojo rw.mojo
mojo rw.mojo  3.14s user 0.09s system 92% cpu 3.493 total
% time python rw.py
python rw.py  8.69s user 0.15s system 126% cpu 6.990 total

Note: only a few drawing primitives are exposed from Cairo right now, but more to come pretty soon!

Setup

For the Python version, create a virtual env and install the requirements.

python3 -m venv venv
source venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

For the Mojo version, please follow the instructions on Modular's website to get Mojo up and running. Make sure you have pycairo installed as well.

Using

Basquiat keeps the same API between Python and Mojo. The idea is to expose a simple Renderer object that keeps the state of the canvas and implements drawing operations.

In Python:

from basquiat.render import Renderer

r = Renderer(1024, 1024)
r.line(0, 512, 512, 512)
r.stroke()

In Mojo:

from basquiat.render import Renderer

let r = Renderer(1024, 1024)
r.line(0, 512, 512, 512)
r.stroke()

Please check the examples (rw.{mojo,py}, lines.{mojo,py}) and also the Mojo and Python implementations in basquiat/.

Studies

Random walk

A couple of random walkers with 100k steps.

Lines

Inspired by Joy Division's Unknown Pleasure album cover.

Noise

Some experiments with uniform and Perlin noise.