Skip to content

oracle-samples/oracle-graph

Repository files navigation

Oracle Graph

Property graphs, RDF graphs, Knowledge graphs, AI-ready analytics, and 80+ built-in algorithms - all native inside Oracle Database. No separate graph engine. No data movement.

Note

Oracle Graph is a native feature of Oracle AI Database 26ai and Oracle Database 19c+. If you already run Oracle Database, you already have Oracle Graph. No extra graph engine is required.

Start Here

Choose the path that matches what you want to do:

I want to... Go to Best for
Start with the current property graph experience property-graph/26ai/get-started/ New users and anyone using Oracle AI Database 26ai
Use property graphs with Oracle Database 19c property-graph/19c/get-started/ Existing 19c users and PGQL-based workflows
Try property graph demos for 26ai property-graph/26ai/demos/ 26ai SQL property graph demos
Try knowledge graph or GraphRAG demos knowledge-graph/demo/ AI and knowledge graph examples
Browse graph algorithm examples property-graph/algorithms/ PageRank, shortest path, centrality, community detection, and related samples
Reuse datasets or helper utilities shared/ Files used by more than one demo or database version

If you are new to Oracle Graph, start with property-graph/26ai/get-started/. Use the 19c path only when you are specifically working with Oracle Database 19c.

What Oracle Graph Gives You

Property Graph

  • Query and traverse graphs using SQL with SQL:2023 property graph syntax (or PGQL for databases 19c and prior).
  • Run 80+ built-in algorithms, including PageRank, Louvain, Dijkstra, betweenness centrality, and more.
  • Use Graph Studio, a browser-based IDE for building, querying, and visualizing graphs.

RDF Graph

  • Create, query, and run entailment on RDF graphs using W3C standards.
  • Use full SPARQL 1.1 support with RDF, RDFS, and OWL ontologies.
  • Build semantic reasoning workflows and knowledge graphs.

AI Integration

  • Use GraphRAG to ground LLM responses with structured graph context.
  • Combine VECTOR_DISTANCE semantic search with graph pattern matching.
  • Build knowledge graphs automatically from unstructured text.

Repository Contents

Folder What It Holds
property-graph/ Property graph examples, organized by database version plus algorithm samples.
knowledge-graph/ Knowledge graph and GraphRAG examples.
rdf/ RDF graph examples and assets.
shared/ Reusable datasets, utilities, and assets shared across examples.

Common Use Cases

Use Case Where To Look
Build a first property graph in 26ai property-graph/26ai/get-started/
Build a first property graph in 19c property-graph/19c/get-started/
Try 26ai property graph demos property-graph/26ai/demos/
Try GraphRAG or knowledge graph demos knowledge-graph/demo/
Browse graph algorithms property-graph/algorithms/
Find reusable datasets or helper scripts shared/
Check property graph version compatibility property-graph/docs/compatibility.md

Folder Guide

Use this folder for property graph examples. It contains version-specific getting-started paths, demos, and graph algorithm samples.

Expected contents:

Use this folder for knowledge graph, AI, and GraphRAG examples.

Use this folder for RDF graph examples and related assets.

Use this folder only for files reused by more than one example, such as datasets, helper scripts, or common assets. If a file belongs to one demo only, keep it inside that demo folder.

Use this folder for repository-level guidance, including compatibility notes and migration guidance between Oracle Database 19c and Oracle AI Database 26ai.

Prerequisites

The examples in this repository may require:

  • Oracle AI Database 26ai or Oracle Database 19c.
  • Oracle Graph components enabled for your database environment.
  • SQLcl, SQL Developer, Graph Studio, or another supported SQL client.
  • Python and Jupyter for notebook-based walkthroughs.
  • Optional: Conda for custom Graph Studio Python interpreter setup.

Check the README inside each folder for version-specific setup steps and dependencies.

Documentation

Repository Validation

Run the validation script before opening a pull request or after moving files:

node scripts/validate-repo.js

The script checks local Markdown links, JSON-based notebooks, cleared Jupyter outputs, required README files, and stray .DS_Store files.

Contributing

Contributions are welcome. Before opening a pull request, please:

  • Keep examples self-contained and reproducible.
  • Include setup notes for any database, Python, or notebook dependencies.
  • Place version-specific property graph examples under property-graph/26ai/ or property-graph/19c/.
  • Place reusable datasets, utilities, and assets under shared/ only when they are used by more than one example.
  • Add expected output or screenshots when the result is visual.
  • Avoid committing credentials, connection strings, wallets, or generated data that should not be shared.

Support

For questions, issues, or demo requests, open a GitHub issue or contact the repository maintainers.

Packages

 
 
 

Contributors