networkxr 0.1.5

Drop-in replacement for NetworkX (Rust-accelerated)

networkXR

[!CAUTION] 🚧 Under Heavy Construction 🚧

This project is in early development. APIs may change without notice. Not yet ready for production use.

Goal

networkXR is a Rust-backed, drop-in replacement for NetworkX — the popular Python library for graph creation, manipulation, and analysis.

By leveraging PyO3 and Rust's performance characteristics, networkXR aims to provide a fully compatible NetworkX API with significantly improved speed for compute-intensive graph operations, while remaining a seamless swap for existing Python codebases.

Why networkXR?

• 🚀 Blazing Fast: Core graphs (Graph, DiGraph) use Rust IndexMap internally for high-performance processing.
• 🤝 Drop-in Replaceable: Change your import networkx as nx to import networkxr as nx. That's it.
• 🎨 Batteries Included: Comes with generators (including fake data), converters, isomorphism checks, and a rich Plotly-native visualization engine.

Installation

pip install networkxr

# With interactive plotting support (optional)
pip install networkxr[plot]

Quick Start

# Just swap the import — everything else stays the same
import networkxr as nx

G = nx.Graph()
G.add_edges_from([(1, 2), (2, 3), (3, 4), (4, 1), (1, 3)])
G.add_node(5, role="isolated")

print(G.number_of_nodes())   # 5
print(G.number_of_edges())   # 5
print(list(G.neighbors(1)))  # [2, 3, 4]

Plotting & Visualization

networkXR natively supports Plotly for interactive, publication-ready graph visualization out-of-the-box (no matplotlib required!).

import networkxr as nx

# Create a beautiful graph topology
G = nx.barbell_graph(5, 1)

# Draw it with custom styling
fig = nx.draw(
    G, 
    node_color="#6366f1", 
    edge_color="#cbd5e1", 
    title="Barbell Graph Visualization", 
    show=False
)
fig.write_html("my_graph.html")

Note: Plotting requires plotly — install with pip install networkxr[plot].

Rich Data Generation

Need a realistic graph for testing? networkXR includes Faker-powered network generation:

import networkxr as nx

# Create a realistic social network graph with user attributes
G = nx.fake_social_network(n=50, p=0.1, seed=42)

for u, data in list(G.nodes(data=True))[:2]:
    print(f"{data['name']} ({data['email']}) works at {data['company']}")

Documentation

For full API reference, examples, and cookbooks, check out our documentation website.