Raphtory GraphQL Schema Explorer

🚀 Raphtory GraphQL Schema Explorer

This Python module provides a FastMCP server that exposes the GraphQL schema information of Raphtory graphs through a set of HTTP resources. It enables LLMs to explore and understand the structure of the graphs, including node attributes and relationship types.

🚀 Quick Start

The Raphtory GraphQL Schema Explorer is a powerful tool that allows you to interact with the GraphQL schema of Raphtory graphs. With this module, you can query graph schemas, obtain detailed information about node attributes and relationship types, and more.

✨ Features

1. Graph Schema Query

• Query the schema of a specific graph by name.
• Get detailed information about node attributes.
• Retrieve unique relationship types and their metadata.
• Optionally include attribute variants.

2. Resources

schema://database

Returns the full GraphQL schema of the database, including:

• All available types.
• Fields for each type.
• Field descriptions.
• Parameters and their default values.

schema://graph_exists/{graph_name}

Verifies the existence of a specific graph in the database by checking the presence of nodes.

schema://{graph_name}/{include_variants}

Returns detailed schema information for a specific graph:

• Node attributes:
  • Attribute keys.
  • Attribute types (e.g., "Str").
  • Optional attribute variants.
• Relationships:
  • Unique relationship types (e.g., "edge_type_1", "edge_type_2", "edge_type_3").

3. Implementation Details

This module uses:

• A running Raphtory GraphQL server (needs to be started).
• httpx for asynchronous HTTP requests, supporting HTTP/2.
• FastMCP to manage resources.
• Persistent connections with a global asynchronous client.
• Context management to properly clean up resources.

💻 Usage Examples

Basic Usage

Query the graph schema:

{
  graph(path: "your_graph_name") {
    schema {
      nodes {
        properties {
          key
          propertyType
        }
      }
    }
    edges {
      list {
        properties {
          keys
          values {
            value
          }
        }
      }
    }
  }
}

Advanced Usage

The above example demonstrates how to query the basic information of a graph schema. You can adjust the query according to your specific needs, such as querying more detailed node attributes or relationship information.

📚 Documentation

Response Format

The schema response includes values, for example:

{
  "data": {
    "graph": {
      "schema": {
        "nodes": [{
          "properties": [
            {"key": "name", "propertyType": "Str"},
            {"key": "type", "propertyType": "Str"}
            // ... other attributes
          ]
        }]
      },
      "relationships": [
        "edge_type_1",
        "edge_type_2",
        "edge_type_3"
      ]
    }
  }
}

Error Handling

This module includes the following error handling:

• Non-existent graphs.
• Invalid queries.
• Connection issues.
• Incorrect response formats.

Dependencies

• FastMCP
• httpx
• Python 3.x
• A running Raphtory GraphQL endpoint (default: http://localhost:1736/)

Configuration

This module uses the default configuration:

• HTTP/2 enabled.
• 10-second timeout.
• Content-Type: application/json

You can also add additional configurations through environment variables, such as authentication.