pringle 0.2.0

Interactive graph and surface plotter with python expressions.

P R I N G L E

Interactive 3D plotting for Python. Define surfaces, vector fields, scatter plots, and dynamical system trajectories using real Python/numpy expressions — then explore them live with sliders and animation.

Install + Run:

pip install pringle

crunch 

Why Pringle?

Real Python — expressions are plain Python/numpy. sin(x)*cos(y), scipy.special.gamma(x), user-defined lambdas — no domain-specific language to learn. Numpy functions are available by name (sin, cos, linspace, random, …) with no prefix required.

Live interactivity — any named scalar assignment (a = 1.0) becomes a slider. Drag it and every dependent expression re-evaluates instantly via a reactive dependency graph, at 60 fps on a background thread.

User controls - Fly through your plot like a video game, using the WASD controls to move around + SPACE/SHIFT to move up and down.

Animate anything — press ▷ on any slider and it loops over its range automatically. Camera orbit and pan stay fluid even while the scene is re-evaluating.

Version-controlled sessions — plot sessions are stored in human-readable YAML files. git diff shows exactly what changed; send a single .yml file to share a visualization with a collaborator.

LLM Compatibility — human readable session files + python-native expressions mean large language models are able to generate sophisticated pringle plots on the fly.

GPU-accelerated rendering — powered by pygfx + wgpu-py. Smooth Phong shading, order-independent transparency, and drop shadows — not matplotlib's software renderer.

Vector fields — native 2D and 3D arrow rendering, auto-detected from array shapes (N, 4) and (N, 6).

Recurrence relations — express ODE trajectories directly: path[n] = path[n-1] + dt * deriv(path[n-1]) with an initial condition sub-cell. Re-evaluates reactively on every slider tick.

Piecewise expressions — z = [f, g, h] with condition sub-cells produces a single unified surface.

Constraint masking — boolean sub-cells clip any surface without touching the expression.

Folders and comments — organize complex sessions; collapsed folders keep the expression panel clean.

Pringle vs. Other Tools

	Pringle	Desmos 3D	matplotlib Axes3D	Plotly
Language	Python / numpy	Custom math notation	Python	Python
Live sliders	✓	✓	—	Requires Dash
GPU rendering	✓ (wgpu)	✓ (WebGL)	—	—
Vector fields	✓	—	—	Limited
Recurrence / ODE paths	✓	Limited	Manual	Manual
scipy / custom functions	✓	—	✓	✓
Version-controllable	✓ (YAML)	—	—	—

Installation

pip install pringle

Requires Python ≥ 3.11.

Running Pringle

# Open a saved session
crunch path/to/session.yml

# Open a tutorial
crunch pringle/examples/tutorials/01_hello_surface.yml

# Alternative pringle alias
pringle pringle/examples/tutorials/03_animation.yml

Expression Language

Cells evaluate as plain Python. Spatial variables x, y, u, v are injected as numpy arrays over the evaluation grid. Output type is determined by the magic variable assigned:

Assignment	Renders as
z = ...	Explicit surface (N, M)
xyz = ...	Parametric surface or curve (3, N, M) or (3, N)
points = ...	Scatter plot (N, 3) or (N, 2)
f(x, y) = ...	Auto-rendered surface + reusable namespace function

Shapes (N, 4) and (N, 6) are auto-detected as 2D and 3D vector fields respectively. They are viewed as N row vectors, with each row having the tail and head coordinates concatenated together.

Tutorials

Step-by-step sessions in pringle/examples/tutorials/:

File	Concept
01_hello_surface.yml	First surface — explicit z = sin(x) * cos(y)
02_sliders.yml	Interactive slider parameters
03_animation.yml	Animated slider — traveling wave
04_parametric.yml	Parametric surface — sphere via xyz
05_constraints.yml	Constraint sub-cells — clipping a surface
06_scatter.yml	Scatter plot — helix point cloud
07_vector_field.yml	2D vector field — gradient of a surface
08_recurrence.yml	Recurrence relation — integrating an ODE

More complex real-world examples live in pringle/examples/.

Further Reading

• design-docs/01-desmos-3d-overview.md — inspiration and comparison with Desmos 3D
• design-docs/03-expression-evaluation.md — expression language, security model, and namespace
• design-docs/04-animation-and-time-variation.md — animation loop and performance model
• design-docs/07-cell-types-and-blocks.md — all cell types: equation, slider, folder, comment; piecewise and constraints
• design-docs/08-visual-styling.md — colors, colormaps, opacity, and render modes
• design-docs/10-session-format.md — full YAML session schema reference
• design-docs/11-recurrence-relations.md — recurrence syntax, execution model, and ODE integration
• design-docs/12-user-input-and-interaction.md — keyboard shortcuts, camera controls, WASD navigation