Introduction

At a glance

pgRDF · RDF, SPARQL, SHACL and OWL reasoning inside PostgreSQL · License MIT · PostgreSQL 18 · Latest release v0.6.34 · Install → · Ten-minute tour →

pgRDF is a PostgreSQL extension, written in Rust, that turns your database into a knowledge-graph engine. With it you can:

• load RDF;
• query and update it with SPARQL 1.1;
• reason over it with OWL 2 RL or RDFS;
• validate it with SHACL;
• lock, fingerprint and export it.

Every operation is a SQL function call, so any PostgreSQL client already works with it.

CREATE EXTENSION pgrdf;

SELECT pgrdf.add_graph('http://example.org/people');

SELECT pgrdf.parse_turtle($$
  @prefix ex:   <http://example.org/> .
  @prefix foaf: <http://xmlns.com/foaf/0.1/> .
  ex:alice foaf:name "Alice" ; foaf:knows ex:bob .
  ex:bob   foaf:name "Bob" .
$$, pgrdf.graph_id('http://example.org/people'));

SELECT * FROM pgrdf.sparql($$
  PREFIX foaf: <http://xmlns.com/foaf/0.1/>
  SELECT ?who ?friend WHERE { ?a foaf:name ?who ; foaf:knows ?b . ?b foaf:name ?friend }
$$);
--  {"who": "Alice", "friend": "Bob"}

Try it in two minutes

With Docker and nothing else, run the extension in a stock postgres:18 container:

VER=0.6.34
ARCH=$(uname -m | sed -e 's/x86_64/amd64/' -e 's/aarch64/arm64/')
curl -fsSL https://github.com/styk-tv/pgRDF/releases/download/v$VER/pgrdf-$VER-pg18-glibc-$ARCH.tar.gz | tar -xz

docker create --name pgrdf -e POSTGRES_PASSWORD=pgrdf -p 5432:5432 \
  postgres:18 -c shared_preload_libraries=pgrdf
docker cp pgrdf-$VER-pg18-glibc-$ARCH/lib/pgrdf.so       pgrdf:/usr/lib/postgresql/18/lib/
docker cp pgrdf-$VER-pg18-glibc-$ARCH/share/extension/.  pgrdf:/usr/share/postgresql/18/extension/
docker start pgrdf

until docker exec pgrdf pg_isready -h localhost -U postgres -q; do sleep 1; done
docker exec -it pgrdf psql -U postgres -c 'CREATE EXTENSION pgrdf'
docker exec -it pgrdf psql -U postgres

Then follow the ten-minute tour. Other routes (your own image, an existing server, Kubernetes, the signed OCI bundle) are on the install page.

What's inside

	What you get	Start here
storage Load	Turtle, N-Triples, TriG, N-Quads, from SQL strings or server files; parallel loaders for very large N-Triples dumps	Storage
search Query	SPARQL 1.1 SELECT / ASK / CONSTRUCT / DESCRIBE, aggregates, property paths, named graphs, full UPDATE; results are JSONB rows you can join with any table	Query
psychology Reason	OWL 2 RL or RDFS materialization, stored beside your data, with a freshness flag	Inference
fact_check Validate	W3C SHACL Core (25/25 on the conformance suite) plus SHACL-SPARQL, with the report as JSONB	Validation
account_tree Manage graphs	Inventory, copy / move / clear / drop, carving by predicate or neighbourhood, write locks, integrity checks	Graphs
verified Identity and export	Canonical graph digests (W3C RDFC-1.0), canonical N-Triples export, a manifest anyone can verify offline	Identity
settings Operate	Standard SQLSTATE error codes, per-query completeness, settings, capability listing, statistics	Errors and diagnostics

Why inside PostgreSQL

	A separate triple store	pgRDF
Where the graph lives	Another server to run	In your PostgreSQL database
Backups, replication, point-in-time recovery	Its own tooling	PostgreSQL's
Users, roles, TLS	Its own	PostgreSQL's
Combining graph and relational data	Export, sync, glue code	One SQL query
Transactions	Separate	Graph changes commit or roll back with the rest of your transaction

Who it's for

• code Application developers: start with the tour and the client guides for Python, Node.js, Go and Rust.
• query_stats Data engineers: see loading, SPARQL and combining with SQL.
• school Ontologists: see reasoning and validation.
• settings Operators: see install, errors and diagnostics and scale.
• build Contributors: see the internals.

Scale

The same extension runs a few triples in a container and very large graphs on one server:

• a complete Wikidata "truthy" dump, 8.2 billion triples, loaded into a single instance;
• the full load → reason → query pipeline at LUBM-500, ending in a 112-million-quad materialized closure.

See Scale and benchmarks.

Get the bits

• Releases: github.com/styk-tv/pgRDF/releases. Archives for PostgreSQL 18 on x86-64 and arm64, with SHA256SUMS.
• Signed OCI bundle: ghcr.io/styk-tv/pgrdf-bundle, with SLSA build provenance you can check with gh attestation verify oci://ghcr.io/styk-tv/pgrdf-bundle:0.6.34 --repo styk-tv/pgRDF.
• Current version and digests:LATEST.md.
• What changed: CHANGELOG.
• Source: github.com/styk-tv/pgRDF.

Next — the ten-minute tour →