anywidget-graph¶

Interactive graph visualization widget for Python notebooks.

Overview¶

anywidget-graph provides interactive graph visualization powered by Sigma.js. Built on the anywidget framework, it works universally across Jupyter, Marimo, VS Code, Colab, and Databricks with support for multiple graph database backends.

Features¶

Multi-Backend Support¶

Backend	Query Languages
Grafeo	GQL, Cypher, SPARQL, Gremlin, GraphQL
Neo4j	Cypher
ArangoDB	AQL
LadybugDB	Cypher
NetworkX	Python API

Query Language Converters¶

Automatically converts query results from:

Cypher (Neo4j, Grafeo)
GQL (ISO standard)
SPARQL (RDF/semantic web)
Gremlin (TinkerPop)
GraphQL (JSON responses)
AQL (ArangoDB)

Interactivity¶

Pan, zoom, and navigate
Click nodes/edges to select
Expand/collapse neighbors
Multi-select paths
Export to PNG, SVG, JSON

Installation¶

uv add anywidget-graph

With optional backends:

uv add "anywidget-graph[networkx,neo4j]"

Quick Start¶

From Grafeo¶

from anywidget_graph import Graph
from grafeo import GrafeoDB

db = GrafeoDB()
db.execute("""
    INSERT (:Person {name: 'Alice'})-[:KNOWS]->(:Person {name: 'Bob'}),
           (:Person {name: 'Bob'})-[:KNOWS]->(:Person {name: 'Carol'})
""")

result = db.execute("MATCH (n)-[r]->(m) RETURN n, r, m")
graph = Graph.from_grafeo(result)
graph

From NetworkX¶

import networkx as nx
from anywidget_graph import Graph

G = nx.karate_club_graph()
widget = Graph.from_networkx(G)
widget

From Edge List¶

from anywidget_graph import Graph

nodes = [
    {"id": "1", "label": "Alice", "group": "person"},
    {"id": "2", "label": "Bob", "group": "person"},
    {"id": "3", "label": "Acme", "group": "company"},
]

edges = [
    {"source": "1", "target": "2", "label": "knows"},
    {"source": "2", "target": "3", "label": "works_at"},
]

widget = Graph(nodes=nodes, edges=edges)
widget

Styling¶

Node Styling¶

widget = Graph(
    nodes=nodes,
    edges=edges,
    node_color_by="group",
    node_size_by="degree",
    node_colors={
        "person": "#6366f1",
        "company": "#22c55e"
    }
)

Layout Algorithms¶

widget = Graph(
    nodes=nodes,
    edges=edges,
    layout="force",  # force, hierarchical, circular, grid
)

Event Handling¶

@widget.on_node_click
def handle_node_click(node_id, node_data):
    print(f"Clicked: {node_data['label']}")

@widget.on_edge_click
def handle_edge_click(edge_id, edge_data):
    print(f"Edge: {edge_data['label']}")

Configuration¶

widget = Graph(
    nodes=nodes,
    edges=edges,

    # Display
    show_labels=True,
    show_edge_labels=False,
    show_toolbar=True,
    dark_mode=True,

    # Layout
    layout="force",

    # Performance
    virtualize=True,  # For 1000+ nodes
)

Database Connections¶

Neo4j¶

from anywidget_graph import Graph
from anywidget_graph.backends import Neo4jBackend

backend = Neo4jBackend(
    uri="bolt://localhost:7687",
    user="neo4j",
    password="password"
)

widget = Graph(
    backend=backend,
    query="MATCH (n)-[r]->(m) RETURN n, r, m LIMIT 100"
)

ArangoDB¶

from anywidget_graph.backends import ArangoDBBackend

backend = ArangoDBBackend(
    host="localhost",
    database="mydb",
    username="root",
    password="password"
)

widget = Graph(
    backend=backend,
    query="FOR v, e IN 1..2 OUTBOUND 'nodes/1' edges RETURN {v, e}"
)

Notebook Compatibility¶

Works in any environment supporting anywidget:

JupyterLab / Jupyter Notebook
Marimo
VS Code Notebooks
Google Colab
Databricks

Requirements¶

Python 3.12+
Modern browser

License¶

Apache-2.0