NetworkX Graph API

A FastAPI application for uploading, storing, and analyzing NetworkX graphs.

Features

• Upload NetworkX graphs in JSON format
• Retrieve comprehensive graph summaries
• Unique ID assignment for each uploaded graph
• Local storage of graph files
• Set a default graph that can be accessed without uploading a file
• Access graphs using the special "default" graph ID
• File-based workflow: Load graphs directly from files in graph_storage/

Installation

# Create and activate virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Running the API

uvicorn main:app --reload --host 0.0.0.0 --port 8000

The API will be available at http://localhost:8000

API Endpoints

POST /upload/

Upload a NetworkX graph JSON file.

Parameters:

• file (required): JSON file containing a NetworkX graph

Response:

{
  "graph_id": "unique-id-for-this-graph",
  "message": "Graph uploaded successfully",
  "filename": "original-filename.json"
}

Example:

curl -X POST -F "file=@sample_graph.json" http://localhost:8000/upload/

POST /load_graph

Load a graph from a file in the graph_storage directory.

Parameters:

• filename (required): Name of the file in graph_storage/ directory

Response:

{
  "graph_id": "unique-id-for-this-graph",
  "message": "Graph loaded from file successfully",
  "filename": "filename.json"
}

Example:

curl -X POST -H "Content-Type: application/json" -d '{"filename": "my_graph.json"}' http://localhost:8000/load_graph

GET /summary/{graph_id}

Get a summary of the properties of a NetworkX graph.

Parameters:

• graph_id (path): The unique ID returned when uploading the graph, or "default" to access the default graph

Response:

{
  "graph_id": "unique-id-for-this-graph",
  "basic_properties": {
    "number_of_nodes": 5,
    "number_of_edges": 7,
    "is_directed": true,
    "is_weakly_connected": true,
    "is_strongly_connected": true,
    "is_connected": null
  },
  "degree_properties": {
    "average_degree": 2.8,
    "degree_centrality": {"1": 0.42857142857142855, "2": 0.42857142857142855, ...}
  },
  "node_properties": {
    "node_count": 5,
    "nodes_with_highest_degree": [[1, 2], [2, 2], [3, 2], [4, 1], [5, 1]]
  },
  "edge_properties": {
    "edge_count": 7,
    "edges_with_highest_weight": [[4, 5, 2.0], [1, 2, 2.5], [3, 4, 3.0], [2, 4, 1.5], [1, 3, 1.0], [5, 1, 1.0], [2, 3, 0.5]]
  }
}

Example:

# Access a specific graph by its ID
curl http://localhost:8000/summary/your-graph-id-here

# Access the default graph
curl http://localhost:8000/summary/default

GET /health/

Health check endpoint to verify the API is running.

Response:

{
  "status": "healthy",
  "graph_count": 1
}

File-based Workflow

The API supports loading graphs directly from files in the graph_storage/ directory. This is useful when you have pre-existing graph files that you want to analyze without uploading them through the API.

Setup Process:

1. Place your graph files in the graph_storage/ directory
2. Load the graph using the POST /load_graph endpoint
3. Access it using any of the standard endpoints with the returned graph_id

Example Workflow:

# 1. Place your graph file in graph_storage/
cp my_graph.json graph_storage/

# 2. Load the graph
curl -X POST -H "Content-Type: application/json" -d '{"filename": "my_graph.json"}' http://localhost:8000/load_graph
# Response: {"graph_id": "abc123...", "message": "Graph loaded from file successfully", ...}

# 3. Access the graph
curl http://localhost:8000/summary/abc123...
# Response: {graph summary...}

Setting a File-based Graph as Default:

# 1. Load the graph from file
curl -X POST -H "Content-Type: application/json" -d '{"filename": "my_graph.json"}' http://localhost:8000/load_graph
# Response: {"graph_id": "abc123...", ...}

# 2. Set it as default
curl -X POST http://localhost:8000/set-default/abc123...
# Response: {"message": "Graph set as default successfully", ...}

# 3. Access the default graph
curl http://localhost:8000/summary/default
# Response: {graph summary...}

Testing

Run the test scripts to create a sample graph and test the API endpoints:

pytest

For details, see /tests/README.md.

Graph Format

The JSON file should contain a NetworkX graph in node-link format, which can be created using:

import networkx as nx
data = nx.node_link_data(G)  # Where G is your NetworkX graph

Project Structure

.
├── main.py                # FastAPI application
├── requirements.txt       # Dependencies
├── README.md              # This file
├── tests/                 # Test suite
└── graph_storage/        # Directory for stored graphs (created automatically)

Dependencies

• FastAPI: Web framework
• Uvicorn: ASGI server
• NetworkX: Graph library

License

This project is open source and available under the MIT License.