Skip to content

sidprasad/spytial

Repository files navigation

sPyTial: Lightweight Diagrams for Structured Python Data

PyPI version Python versions License: MIT CI Docs

pip install spytial-diagramming

Docs: https://sidprasad.github.io/spytial/ — start with Getting Started for install, badges, and your first diagram.


Sometimes you just want to see your data.

You're working with a tree, a graph, a recursive object — maybe an AST, a neural network, or a symbolic term. You don't need an interactive dashboard or a production-grade visualization system. You just need a diagram that lays it out clearly so you can understand what's going on.

That's what sPyTial is for. It is designed for developers, educators, and researchers who work with structured data and need to make that structure visible — to themselves or to others — with minimal effort.

Why spatial layout

Spatial arrangement helps people understand structure: when elements are grouped, aligned, and oriented meaningfully, patterns and errors become easier to see. sPyTial gives you that layout by default — the diagram reflects how the parts are connected, not how they happen to be stored.

You get:

  • a box-and-arrow diagram that shows the shape of your data
  • a layout driven by declarative constraints (orientation, align, cyclic, group)
  • a tool that flags when a constraint can't be satisfied

Quick start

import spytial

data = {
    "name": "root",
    "children": [
        {"value": 1},
        {"value": 2},
        {"value": 3},
    ],
}

# Opens in a browser tab, or inline in a Jupyter notebook.
spytial.diagram(data)

# Or save to a file:
spytial.diagram(data, method="file")

For stepping through sequences of states, custom relationalizers, and annotation-driven layouts, see the docs.

Related projects

  • spytial-clrs — a Jupyter notebook collection that implements the data structures from the CLRS algorithms textbook (Cormen, Leiserson, Rivest, Stein) using sPyTial: heaps, linked lists, hash tables, BST / red-black / B / van Emde Boas trees, disjoint-set forests, graphs with MST and SCC views, and more. The best place to see sPyTial on realistic structures.
  • spytial-core — the browser-side rendering engine sPyTial uses under the hood. Shared across all sPyTial language hosts (Python, Rust, Pyret, Lean).

License

MIT — see LICENSE.

About

sPyTial: Lightweight Diagrams for Structured Python Data

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors