psychologyReason

Materialize the implicit consequences of a graph — subclass, subproperty, equivalence, inverse, transitive — and write them back as queryable quads. Reason is single-threaded, so it runs over a graph sized to your hardware.

What it is

Reason forward-chains the entailment closure under a chosen profile and writes every inferred triple back into the same partition as is_inferred = TRUE rows. After Reason, a Query sees both asserted and entailed triples. The call is idempotent — prior inferred rows are dropped and re-derived each run — and it refreshes planner statistics automatically.

How you run it

pgrdf.materialize(graph_id BIGINT, profile TEXT DEFAULT 'owl-rl') → JSONB

The profile selects the closure:

• 'owl-rl' (default) — full OWL 2 RL forward-chaining.
• 'rdfs' — RDFS closures only.

An unknown profile raises an error — there is no silent fallback. See Pillar 3 — Materialization for the rule set.

Where it sits in a chain

After Seal, before Validate and Query. Reason is the verb that decides whether you need to carve: if the graph is larger than one backend can close over, you Carve a right-sized slice and reason over that.

Scaling class — single-threaded

Reason runs on one backend. OWL-RL materialization is single-thread-bound upstream in the reasonable reasoner (issue #1). This is the binding constraint of the operating model: you do not reason over the full 8.2 B source graph on ordinary hardware — you reason over a graph sized to your box. See Reasoning at scale.

Example

-- ex:alice a ex:Engineer ; ex:Engineer ⊑ ex:Person ⊑ ex:Agent
SELECT pgrdf.materialize(100);
--  → {"base_triples": 3, "inferred_triples_written": 11, ...}

-- the 2-hop entailment is now queryable
SELECT * FROM pgrdf.sparql(
  'PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
   PREFIX ex: <http://example.com/>
   SELECT ?c WHERE { ex:alice rdf:type ?c }');
--  → ex:Engineer (base), ex:Person (inferred), ex:Agent (inferred)

See also

• Pillar 3 — Materialization — the OWL 2 RL + RDFS rule set.
• Reasoning at scale — why reasoning runs on a sized slice.
• Carve — the roadmap verb that right-sizes a graph for Reason.