@ruvector/graph Node

@ruvector/graph-node

Native Node.js bindings for the RuVector Graph Database with hypergraph support, Cypher query engine, graph traversal algorithms, and persistent storage for building knowledge graphs and relationship-heavy applications.

Quick Command Reference

Installation

Hub install (recommended): npx ruvector@latest includes this package. Standalone: npx @ruvector/graph-node@latest See Installation Guide for the full ecosystem.

Core API

GraphDB Constructor

import { GraphDB } from '@ruvector/graph-node';

const gdb = new GraphDB({
  persistPath: './graph-data',
  enableCypher: true,
  enableHypergraph: true,
});

Constructor Options:

Vertex Operations

// Add vertex with label and properties
const id = await gdb.addVertex('user', { name: 'Alice', age: 30 });

// Get vertex
const vertex = await gdb.getVertex(id);

// Update vertex properties
await gdb.updateVertex(id, { age: 31 });

// Delete vertex (and all connected edges)
await gdb.deleteVertex(id);

// List vertices by label
const users = await gdb.getVerticesByLabel('user');

Edge Operations

// Add edge with label and properties
const edgeId = await gdb.addEdge('user:1', 'user:2', 'KNOWS', { since: 2024 });

// Get edge
const edge = await gdb.getEdge(edgeId);

// Delete edge
await gdb.deleteEdge(edgeId);

// Get edges between vertices
const edges = await gdb.getEdges('user:1', 'user:2');

Cypher Queries

// Create
await gdb.query("CREATE (n:Person {name: 'Alice', age: 30})");

// Match and return
const result = await gdb.query("MATCH (n:Person) WHERE n.age > 25 RETURN n");

// Relationships
await gdb.query("MATCH (a:Person {name:'Alice'}), (b:Person {name:'Bob'}) CREATE (a)-[:KNOWS]->(b)");

// Pattern matching
const result = await gdb.query("MATCH (a)-[:KNOWS]->(b)-[:KNOWS]->(c) RETURN DISTINCT c.name");

// With parameters
const result = await gdb.query("MATCH (n:Person {name: $name}) RETURN n", { name: 'Alice' });

Graph Algorithms

// Neighbors
const neighbors = await gdb.neighbors('user:1', { direction: 'out', depth: 2 });

// Shortest path
const path = await gdb.shortestPath('user:1', 'user:5');

// All paths
const paths = await gdb.allPaths('user:1', 'user:5', { maxDepth: 5 });

// PageRank
const ranks = await gdb.pageRank({ iterations: 20, damping: 0.85 });

// Connected components
const components = await gdb.connectedComponents();

// Degree centrality
const centrality = await gdb.degreeCentrality();

Common Patterns

Social Network Graph

const gdb = new GraphDB({ persistPath: './social-graph' });
await gdb.addVertex('user', { name: 'Alice' });
await gdb.addVertex('user', { name: 'Bob' });
await gdb.addEdge('user:1', 'user:2', 'FOLLOWS');
const followers = await gdb.query("MATCH (n)-[:FOLLOWS]->(u:user {name:'Bob'}) RETURN n.name");

Knowledge Graph for RAG

await gdb.query("CREATE (t:Topic {name: 'Machine Learning'})");
await gdb.query("CREATE (c:Concept {name: 'Neural Networks'})");
await gdb.query("MATCH (t:Topic {name:'Machine Learning'}), (c:Concept {name:'Neural Networks'}) CREATE (t)-[:CONTAINS]->(c)");
const related = await gdb.query("MATCH (t:Topic)-[:CONTAINS]->(c) RETURN c.name");

Recommendation Engine

const recommendations = await gdb.query(`
  MATCH (u:User {id: $userId})-[:PURCHASED]->(p)<-[:PURCHASED]-(other)-[:PURCHASED]->(rec)
  WHERE NOT (u)-[:PURCHASED]->(rec)
  RETURN rec.name, count(*) AS score ORDER BY score DESC LIMIT 5
`, { userId: 'user-123' });

Key Options

RAN DDD Context

Bounded Context: Data Infrastructure

References

• API reference: See references/commands.md
• Full README
• npm