Loading Ontologies

curl -X POST http://localhost:8080/ontology \
  -H "Content-Type: text/turtle" \
  --data-binary @your-ontology.ttl

This single request parses, validates, compiles, applies the schema, persists the ontology, initializes the reasoning engine, and triggers retroactive materialization.

# Load an ontology into a running cluster
dgraph ontology load your-ontology.ttl --alpha http://localhost:8080

# Validate without applying
dgraph ontology validate your-ontology.ttl --alpha http://localhost:8080

Test your ontology without side effects:

curl -X POST "http://localhost:8080/ontology?validate=true" \
  -H "Content-Type: text/turtle" \
  --data-binary @your-ontology.ttl

Returns the compiled schema and any validation errors without applying changes.

• Turtle syntax — valid Turtle format
• OWL2 RL compliance — all constructs within the RL profile
• Circular subclass detection — A subClassOf B subClassOf A is rejected
• Conflicting characteristics — a property can’t be both symmetric and asymmetric
• Undefined references — domain/range referencing non-existent classes

The raw Turtle data is stored in the database as an OWLGraphMeta node. On Alpha restart, the ontology is automatically reloaded — no manual intervention needed.

Loading a new ontology replaces the current one. The process:

1. New ontology is parsed and validated
2. New schema is compiled and applied
3. Old reasoning engine is replaced with new one
4. Retroactive materialization runs for existing data

Caution: Ontology updates can trigger significant retroactive materialization if the class hierarchy changes. Test updates in a staging environment first.

• Validate first — always use ?validate=true before applying to production
• Version your ontologies — keep ontology files in version control
• Start simple — begin with core classes, add complexity incrementally
• Test with data — load sample data and run queries to verify behavior