ruviz 0.4.12

Python bindings and notebook widget support for ruviz

ruviz for Python

ruviz for Python wraps the Rust plotting runtime with a fluent Python API, static export helpers, native desktop show(), and notebook widget support.

Install

pip install ruviz

If you want pandas or Polars dataframe inputs, install the optional dataframe extras:

pip install "ruviz[dataframes]"

Quick Start

import numpy as np
import ruviz

x = np.linspace(0.0, 4.0, 50)
y = x**2

(
    ruviz.plot()
    .line(x, y)
    .title("Quadratic")
    .xlabel("x")
    .ylabel("y = x^2")
    .save("quadratic.png")
)

Notebook and Desktop Usage

• In Jupyter, plot.show() displays a static PNG in the cell output.
• Use plot.widget() when you want the synced WASM-backed notebook widget.
• plot.size_px(width, height) also controls the widget's displayed size and aspect ratio.
• Without size_px(...), the widget uses the default PNG size (640x480) and shrinks proportionally if the notebook column is narrower.
• Drag the widget's bottom-right handle to resize the display freely.
• Hold Shift or Ctrl while dragging the handle to preserve the current aspect ratio.
• In the widget, right click opens the export menu and right drag performs box zoom.
• Outside notebooks, plot.show() opens the native interactive window.
• The published Linux wheel focuses on static rendering and notebook widgets. Install from source on Linux if you need the native desktop plot.show() window.
• plot.render_png() and plot.render_svg() return in-memory export data.

Reactive Notebook Data

Use ruviz.observable(...) for notebook-driven updates that keep widget state in sync:

import numpy as np
import ruviz

x = np.linspace(0.0, 6.0, 200)
y = ruviz.observable(np.sin(x))

plot = ruviz.plot().line(x, y).title("Live Sine Wave")
widget = plot.widget()

ObservableSeries supports elementwise arithmetic and NumPy ufuncs. Derived observables stay live until you write to them directly:

from copy import deepcopy
import numpy as np

scaled = np.sin(y * 2.0 + 0.25)
template = deepcopy(plot)

deepcopy(plot) creates an independent live copy with fresh observables, while plot.clone() remains a static snapshot copy.

Documentation

• Python docs source: python/docs/
• Python examples: python/examples/
• Root project README: https://github.com/Ameyanagi/ruviz/blob/main/README.md

Contributor Workflow

cd python
uv sync
uv run maturin develop
uv run python scripts/generate_gallery.py
uv run mkdocs serve

Rebuild the bundled widget frontend from the repository root when you change the web SDK or packages/ruviz-web/src/python-widget.ts:

bun run build:python-widget