apunkt/trustgraph
1
0
Fork 0
mirror of https://github.com/trustgraph-ai/trustgraph.git synced 2026-07-09 05:12:12 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

Changed schema for Value -> Term, majorly breaking change (#622)

Browse source 
* Changed schema for Value -> Term, majorly breaking change

* Following the schema change, Value -> Term into all processing

* Updated Cassandra for g, p, s, o index patterns (7 indexes)

* Reviewed and updated all tests

* Neo4j, Memgraph and FalkorDB remain broken, will look at once settled down
This commit is contained in:
cybermaggedon 2026-01-27 13:48:08 +00:00 • committed by GitHub
parent e061f2c633
commit cf0daedefa
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
86 changed files with 2458 additions and 1764 deletions
Download patch file Download diff file Expand all files Collapse all files

76
trustgraph-flow/trustgraph/extract/kg/agent/extract.py
View file

@ -3,7 +3,7 @@ import json
import urllib.parse
import logging
from ....schema import Chunk, Triple, Triples, Metadata, Value
from ....schema import Chunk, Triple, Triples, Metadata, Term, IRI, LITERAL
from ....schema import EntityContext, EntityContexts
from ....rdf import TRUSTGRAPH_ENTITIES, RDF_LABEL, SUBJECT_OF, DEFINITION
@ -253,32 +253,32 @@ class Processor(FlowProcessor):
for defn in definitions:
entity_uri = self.to_uri(defn["entity"])
# Add entity label
triples.append(Triple(
s = Value(value=entity_uri, is_uri=True),
p = Value(value=RDF_LABEL, is_uri=True),
o = Value(value=defn["entity"], is_uri=False),
s = Term(type=IRI, iri=entity_uri),
p = Term(type=IRI, iri=RDF_LABEL),
o = Term(type=LITERAL, value=defn["entity"]),
))
# Add definition
triples.append(Triple(
s = Value(value=entity_uri, is_uri=True),
p = Value(value=DEFINITION, is_uri=True),
o = Value(value=defn["definition"], is_uri=False),
s = Term(type=IRI, iri=entity_uri),
p = Term(type=IRI, iri=DEFINITION),
o = Term(type=LITERAL, value=defn["definition"]),
))
# Add subject-of relationship to document
if metadata.id:
triples.append(Triple(
s = Value(value=entity_uri, is_uri=True),
p = Value(value=SUBJECT_OF, is_uri=True),
o = Value(value=metadata.id, is_uri=True),
s = Term(type=IRI, iri=entity_uri),
p = Term(type=IRI, iri=SUBJECT_OF),
o = Term(type=IRI, iri=metadata.id),
))
# Create entity context for embeddings
entity_contexts.append(EntityContext(
entity=Value(value=entity_uri, is_uri=True),
entity=Term(type=IRI, iri=entity_uri),
context=defn["definition"]
))
@ -288,61 +288,61 @@ class Processor(FlowProcessor):
subject_uri = self.to_uri(rel["subject"])
predicate_uri = self.to_uri(rel["predicate"])
subject_value = Value(value=subject_uri, is_uri=True)
predicate_value = Value(value=predicate_uri, is_uri=True)
subject_value = Term(type=IRI, iri=subject_uri)
predicate_value = Term(type=IRI, iri=predicate_uri)
if rel.get("object-entity", True):
object_uri = self.to_uri(rel["object"])
object_value = Value(value=object_uri, is_uri=True)
object_value = Term(type=IRI, iri=object_uri)
else:
object_value = Value(value=rel["object"], is_uri=False)
object_value = Term(type=LITERAL, value=rel["object"])
# Add subject and predicate labels
triples.append(Triple(
s = subject_value,
p = Value(value=RDF_LABEL, is_uri=True),
o = Value(value=rel["subject"], is_uri=False),
p = Term(type=IRI, iri=RDF_LABEL),
o = Term(type=LITERAL, value=rel["subject"]),
))
triples.append(Triple(
s = predicate_value,
p = Value(value=RDF_LABEL, is_uri=True),
o = Value(value=rel["predicate"], is_uri=False),
p = Term(type=IRI, iri=RDF_LABEL),
o = Term(type=LITERAL, value=rel["predicate"]),
))
# Handle object (entity vs literal)
if rel.get("object-entity", True):
triples.append(Triple(
s = object_value,
p = Value(value=RDF_LABEL, is_uri=True),
o = Value(value=rel["object"], is_uri=True),
p = Term(type=IRI, iri=RDF_LABEL),
o = Term(type=LITERAL, value=rel["object"]),
))
# Add the main relationship triple
triples.append(Triple(
s = subject_value,
p = predicate_value,
o = object_value
))
# Add subject-of relationships to document
if metadata.id:
triples.append(Triple(
s = subject_value,
p = Value(value=SUBJECT_OF, is_uri=True),
o = Value(value=metadata.id, is_uri=True),
p = Term(type=IRI, iri=SUBJECT_OF),
o = Term(type=IRI, iri=metadata.id),
))
triples.append(Triple(
s = predicate_value,
p = Value(value=SUBJECT_OF, is_uri=True),
o = Value(value=metadata.id, is_uri=True),
p = Term(type=IRI, iri=SUBJECT_OF),
o = Term(type=IRI, iri=metadata.id),
))
if rel.get("object-entity", True):
triples.append(Triple(
s = object_value,
p = Value(value=SUBJECT_OF, is_uri=True),
o = Value(value=metadata.id, is_uri=True),
p = Term(type=IRI, iri=SUBJECT_OF),
o = Term(type=IRI, iri=metadata.id),
))
return triples, entity_contexts

16
trustgraph-flow/trustgraph/extract/kg/definitions/extract.py
View file

@ -9,7 +9,7 @@ import json
import urllib.parse
import logging
from .... schema import Chunk, Triple, Triples, Metadata, Value
from .... schema import Chunk, Triple, Triples, Metadata, Term, IRI, LITERAL
# Module logger
logger = logging.getLogger(__name__)
@ -20,9 +20,9 @@ from .... rdf import TRUSTGRAPH_ENTITIES, DEFINITION, RDF_LABEL, SUBJECT_OF
from .... base import FlowProcessor, ConsumerSpec, ProducerSpec
from .... base import PromptClientSpec
DEFINITION_VALUE = Value(value=DEFINITION, is_uri=True)
RDF_LABEL_VALUE = Value(value=RDF_LABEL, is_uri=True)
SUBJECT_OF_VALUE = Value(value=SUBJECT_OF, is_uri=True)
DEFINITION_VALUE = Term(type=IRI, iri=DEFINITION)
RDF_LABEL_VALUE = Term(type=IRI, iri=RDF_LABEL)
SUBJECT_OF_VALUE = Term(type=IRI, iri=SUBJECT_OF)
default_ident = "kg-extract-definitions"
default_concurrency = 1
@ -142,13 +142,13 @@ class Processor(FlowProcessor):
s_uri = self.to_uri(s)
s_value = Value(value=str(s_uri), is_uri=True)
o_value = Value(value=str(o), is_uri=False)
s_value = Term(type=IRI, iri=str(s_uri))
o_value = Term(type=LITERAL, value=str(o))
triples.append(Triple(
s=s_value,
p=RDF_LABEL_VALUE,
o=Value(value=s, is_uri=False),
o=Term(type=LITERAL, value=s),
))
triples.append(Triple(
@ -158,7 +158,7 @@ class Processor(FlowProcessor):
triples.append(Triple(
s=s_value,
p=SUBJECT_OF_VALUE,
o=Value(value=v.metadata.id, is_uri=True)
o=Term(type=IRI, iri=v.metadata.id)
))
ec = EntityContext(

112
trustgraph-flow/trustgraph/extract/kg/ontology/extract.py
View file

@ -8,7 +8,7 @@ import logging
import asyncio
from typing import List, Dict, Any, Optional
from .... schema import Chunk, Triple, Triples, Metadata, Value
from .... schema import Chunk, Triple, Triples, Metadata, Term, IRI, LITERAL
from .... schema import EntityContext, EntityContexts
from .... schema import PromptRequest, PromptResponse
from .... rdf import TRUSTGRAPH_ENTITIES, RDF_TYPE, RDF_LABEL, DEFINITION
@ -39,6 +39,14 @@ URI_PREFIXES = {
}
def make_term(v, is_uri):
"""Helper to create Term from value and is_uri flag."""
if is_uri:
return Term(type=IRI, iri=v)
else:
return Term(type=LITERAL, value=v)
class Processor(FlowProcessor):
"""Main OntoRAG extraction processor."""
@ -446,9 +454,9 @@ class Processor(FlowProcessor):
is_object_uri = False
# Create Triple object with expanded URIs
s_value = Value(value=subject_uri, is_uri=True)
p_value = Value(value=predicate_uri, is_uri=True)
o_value = Value(value=object_uri, is_uri=is_object_uri)
s_value = make_term(subject_uri, is_uri=True)
p_value = make_term(predicate_uri, is_uri=True)
o_value = make_term(object_uri, is_uri=is_object_uri)
validated_triples.append(Triple(
s=s_value,
@ -609,9 +617,9 @@ class Processor(FlowProcessor):
# rdf:type owl:Class
ontology_triples.append(Triple(
s=Value(value=class_uri, is_uri=True),
p=Value(value="http://www.w3.org/1999/02/22-rdf-syntax-ns#type", is_uri=True),
o=Value(value="http://www.w3.org/2002/07/owl#Class", is_uri=True)
s=make_term(class_uri, is_uri=True),
p=make_term("http://www.w3.org/1999/02/22-rdf-syntax-ns#type", is_uri=True),
o=make_term("http://www.w3.org/2002/07/owl#Class", is_uri=True)
))
# rdfs:label (stored as 'labels' in OntologyClass.__dict__)
@ -620,18 +628,18 @@ class Processor(FlowProcessor):
if isinstance(labels, list) and labels:
label_val = labels[0].get('value', class_id) if isinstance(labels[0], dict) else str(labels[0])
ontology_triples.append(Triple(
s=Value(value=class_uri, is_uri=True),
p=Value(value=RDF_LABEL, is_uri=True),
o=Value(value=label_val, is_uri=False)
s=make_term(class_uri, is_uri=True),
p=make_term(RDF_LABEL, is_uri=True),
o=make_term(label_val, is_uri=False)
))
# rdfs:comment (stored as 'comment' in OntologyClass.__dict__)
if isinstance(class_def, dict) and 'comment' in class_def and class_def['comment']:
comment = class_def['comment']
ontology_triples.append(Triple(
s=Value(value=class_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#comment", is_uri=True),
o=Value(value=comment, is_uri=False)
s=make_term(class_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#comment", is_uri=True),
o=make_term(comment, is_uri=False)
))
# rdfs:subClassOf (stored as 'subclass_of' in OntologyClass.__dict__)
@ -648,9 +656,9 @@ class Processor(FlowProcessor):
parent_uri = f"https://trustgraph.ai/ontology/{ontology_subset.ontology_id}#{parent}"
ontology_triples.append(Triple(
s=Value(value=class_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#subClassOf", is_uri=True),
o=Value(value=parent_uri, is_uri=True)
s=make_term(class_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#subClassOf", is_uri=True),
o=make_term(parent_uri, is_uri=True)
))
# Generate triples for object properties
@ -663,9 +671,9 @@ class Processor(FlowProcessor):
# rdf:type owl:ObjectProperty
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/1999/02/22-rdf-syntax-ns#type", is_uri=True),
o=Value(value="http://www.w3.org/2002/07/owl#ObjectProperty", is_uri=True)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/1999/02/22-rdf-syntax-ns#type", is_uri=True),
o=make_term("http://www.w3.org/2002/07/owl#ObjectProperty", is_uri=True)
))
# rdfs:label (stored as 'labels' in OntologyProperty.__dict__)
@ -674,18 +682,18 @@ class Processor(FlowProcessor):
if isinstance(labels, list) and labels:
label_val = labels[0].get('value', prop_id) if isinstance(labels[0], dict) else str(labels[0])
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value=RDF_LABEL, is_uri=True),
o=Value(value=label_val, is_uri=False)
s=make_term(prop_uri, is_uri=True),
p=make_term(RDF_LABEL, is_uri=True),
o=make_term(label_val, is_uri=False)
))
# rdfs:comment (stored as 'comment' in OntologyProperty.__dict__)
if isinstance(prop_def, dict) and 'comment' in prop_def and prop_def['comment']:
comment = prop_def['comment']
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#comment", is_uri=True),
o=Value(value=comment, is_uri=False)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#comment", is_uri=True),
o=make_term(comment, is_uri=False)
))
# rdfs:domain (stored as 'domain' in OntologyProperty.__dict__)
@ -702,9 +710,9 @@ class Processor(FlowProcessor):
domain_uri = f"https://trustgraph.ai/ontology/{ontology_subset.ontology_id}#{domain}"
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#domain", is_uri=True),
o=Value(value=domain_uri, is_uri=True)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#domain", is_uri=True),
o=make_term(domain_uri, is_uri=True)
))
# rdfs:range (stored as 'range' in OntologyProperty.__dict__)
@ -721,9 +729,9 @@ class Processor(FlowProcessor):
range_uri = f"https://trustgraph.ai/ontology/{ontology_subset.ontology_id}#{range_val}"
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#range", is_uri=True),
o=Value(value=range_uri, is_uri=True)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#range", is_uri=True),
o=make_term(range_uri, is_uri=True)
))
# Generate triples for datatype properties
@ -736,9 +744,9 @@ class Processor(FlowProcessor):
# rdf:type owl:DatatypeProperty
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/1999/02/22-rdf-syntax-ns#type", is_uri=True),
o=Value(value="http://www.w3.org/2002/07/owl#DatatypeProperty", is_uri=True)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/1999/02/22-rdf-syntax-ns#type", is_uri=True),
o=make_term("http://www.w3.org/2002/07/owl#DatatypeProperty", is_uri=True)
))
# rdfs:label (stored as 'labels' in OntologyProperty.__dict__)
@ -747,18 +755,18 @@ class Processor(FlowProcessor):
if isinstance(labels, list) and labels:
label_val = labels[0].get('value', prop_id) if isinstance(labels[0], dict) else str(labels[0])
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value=RDF_LABEL, is_uri=True),
o=Value(value=label_val, is_uri=False)
s=make_term(prop_uri, is_uri=True),
p=make_term(RDF_LABEL, is_uri=True),
o=make_term(label_val, is_uri=False)
))
# rdfs:comment (stored as 'comment' in OntologyProperty.__dict__)
if isinstance(prop_def, dict) and 'comment' in prop_def and prop_def['comment']:
comment = prop_def['comment']
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#comment", is_uri=True),
o=Value(value=comment, is_uri=False)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#comment", is_uri=True),
o=make_term(comment, is_uri=False)
))
# rdfs:domain (stored as 'domain' in OntologyProperty.__dict__)
@ -775,9 +783,9 @@ class Processor(FlowProcessor):
domain_uri = f"https://trustgraph.ai/ontology/{ontology_subset.ontology_id}#{domain}"
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#domain", is_uri=True),
o=Value(value=domain_uri, is_uri=True)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#domain", is_uri=True),
o=make_term(domain_uri, is_uri=True)
))
# rdfs:range (datatype)
@ -790,9 +798,9 @@ class Processor(FlowProcessor):
range_uri = range_val
ontology_triples.append(Triple(
s=Value(value=prop_uri, is_uri=True),
p=Value(value="http://www.w3.org/2000/01/rdf-schema#range", is_uri=True),
o=Value(value=range_uri, is_uri=True)
s=make_term(prop_uri, is_uri=True),
p=make_term("http://www.w3.org/2000/01/rdf-schema#range", is_uri=True),
o=make_term(range_uri, is_uri=True)
))
logger.info(f"Generated {len(ontology_triples)} triples describing ontology elements")
@ -814,9 +822,9 @@ class Processor(FlowProcessor):
entity_data = {} # subject_uri -> {labels: [], definitions: []}
for triple in triples:
subject_uri = triple.s.value
predicate_uri = triple.p.value
object_val = triple.o.value
subject_uri = triple.s.iri if triple.s.type == IRI else triple.s.value
predicate_uri = triple.p.iri if triple.p.type == IRI else triple.p.value
object_val = triple.o.value if triple.o.type == LITERAL else triple.o.iri
# Initialize entity data if not exists
if subject_uri not in entity_data:
@ -824,12 +832,12 @@ class Processor(FlowProcessor):
# Collect labels (rdfs:label)
if predicate_uri == RDF_LABEL:
if not triple.o.is_uri: # Labels are literals
if triple.o.type == LITERAL: # Labels are literals
entity_data[subject_uri]['labels'].append(object_val)
# Collect definitions (skos:definition, schema:description)
elif predicate_uri == DEFINITION or predicate_uri == "https://schema.org/description":
if not triple.o.is_uri:
if triple.o.type == LITERAL:
entity_data[subject_uri]['definitions'].append(object_val)
# Build EntityContext objects
@ -848,7 +856,7 @@ class Processor(FlowProcessor):
if context_parts:
context_text = ". ".join(context_parts)
entity_contexts.append(EntityContext(
entity=Value(value=subject_uri, is_uri=True),
entity=make_term(subject_uri, is_uri=True),
context=context_text
))

26
trustgraph-flow/trustgraph/extract/kg/ontology/triple_converter.py
View file

@ -8,7 +8,7 @@ with full URIs and correct is_uri flags.
import logging
from typing import List, Optional
from .... schema import Triple, Value
from .... schema import Triple, Term, IRI, LITERAL
from .... rdf import RDF_TYPE, RDF_LABEL
from .simplified_parser import Entity, Relationship, Attribute, ExtractionResult
@ -87,17 +87,17 @@ class TripleConverter:
# Generate type triple: entity rdf:type ClassURI
type_triple = Triple(
s=Value(value=entity_uri, is_uri=True),
p=Value(value=RDF_TYPE, is_uri=True),
o=Value(value=class_uri, is_uri=True)
s=Term(type=IRI, iri=entity_uri),
p=Term(type=IRI, iri=RDF_TYPE),
o=Term(type=IRI, iri=class_uri)
)
triples.append(type_triple)
# Generate label triple: entity rdfs:label "entity name"
label_triple = Triple(
s=Value(value=entity_uri, is_uri=True),
p=Value(value=RDF_LABEL, is_uri=True),
o=Value(value=entity.entity, is_uri=False) # Literal!
s=Term(type=IRI, iri=entity_uri),
p=Term(type=IRI, iri=RDF_LABEL),
o=Term(type=LITERAL, value=entity.entity) # Literal!
)
triples.append(label_triple)
@ -131,9 +131,9 @@ class TripleConverter:
# Generate triple: subject property object
return Triple(
s=Value(value=subject_uri, is_uri=True),
p=Value(value=property_uri, is_uri=True),
o=Value(value=object_uri, is_uri=True)
s=Term(type=IRI, iri=subject_uri),
p=Term(type=IRI, iri=property_uri),
o=Term(type=IRI, iri=object_uri)
)
def convert_attribute(self, attribute: Attribute) -> Optional[Triple]:
@ -159,9 +159,9 @@ class TripleConverter:
# Generate triple: entity property "literal value"
return Triple(
s=Value(value=entity_uri, is_uri=True),
p=Value(value=property_uri, is_uri=True),
o=Value(value=attribute.value, is_uri=False) # Literal!
s=Term(type=IRI, iri=entity_uri),
p=Term(type=IRI, iri=property_uri),
o=Term(type=LITERAL, value=attribute.value) # Literal!
)
def _get_class_uri(self, class_id: str) -> Optional[str]:

26
trustgraph-flow/trustgraph/extract/kg/relationships/extract.py
View file

@ -13,15 +13,15 @@ import urllib.parse
logger = logging.getLogger(__name__)
from .... schema import Chunk, Triple, Triples
from .... schema import Metadata, Value
from .... schema import Metadata, Term, IRI, LITERAL
from .... schema import PromptRequest, PromptResponse
from .... rdf import RDF_LABEL, TRUSTGRAPH_ENTITIES, SUBJECT_OF
from .... base import FlowProcessor, ConsumerSpec, ProducerSpec
from .... base import PromptClientSpec
RDF_LABEL_VALUE = Value(value=RDF_LABEL, is_uri=True)
SUBJECT_OF_VALUE = Value(value=SUBJECT_OF, is_uri=True)
RDF_LABEL_VALUE = Term(type=IRI, iri=RDF_LABEL)
SUBJECT_OF_VALUE = Term(type=IRI, iri=SUBJECT_OF)
default_ident = "kg-extract-relationships"
default_concurrency = 1
@ -127,16 +127,16 @@ class Processor(FlowProcessor):
if o is None: continue
s_uri = self.to_uri(s)
s_value = Value(value=str(s_uri), is_uri=True)
s_value = Term(type=IRI, iri=str(s_uri))
p_uri = self.to_uri(p)
p_value = Value(value=str(p_uri), is_uri=True)
p_value = Term(type=IRI, iri=str(p_uri))
if rel["object-entity"]:
if rel["object-entity"]:
o_uri = self.to_uri(o)
o_value = Value(value=str(o_uri), is_uri=True)
o_value = Term(type=IRI, iri=str(o_uri))
else:
o_value = Value(value=str(o), is_uri=False)
o_value = Term(type=LITERAL, value=str(o))
triples.append(Triple(
s=s_value,
@ -148,14 +148,14 @@ class Processor(FlowProcessor):
triples.append(Triple(
s=s_value,
p=RDF_LABEL_VALUE,
o=Value(value=str(s), is_uri=False)
o=Term(type=LITERAL, value=str(s))
))
# Label for p
triples.append(Triple(
s=p_value,
p=RDF_LABEL_VALUE,
o=Value(value=str(p), is_uri=False)
o=Term(type=LITERAL, value=str(p))
))
if rel["object-entity"]:
@ -163,14 +163,14 @@ class Processor(FlowProcessor):
triples.append(Triple(
s=o_value,
p=RDF_LABEL_VALUE,
o=Value(value=str(o), is_uri=False)
o=Term(type=LITERAL, value=str(o))
))
# 'Subject of' for s
triples.append(Triple(
s=s_value,
p=SUBJECT_OF_VALUE,
o=Value(value=v.metadata.id, is_uri=True)
o=Term(type=IRI, iri=v.metadata.id)
))
if rel["object-entity"]:
@ -178,7 +178,7 @@ class Processor(FlowProcessor):
triples.append(Triple(
s=o_value,
p=SUBJECT_OF_VALUE,
o=Value(value=v.metadata.id, is_uri=True)
o=Term(type=IRI, iri=v.metadata.id)
))
await self.emit_triples(

8
trustgraph-flow/trustgraph/extract/kg/topics/extract.py
View file

@ -11,7 +11,7 @@ import logging
# Module logger
logger = logging.getLogger(__name__)
from .... schema import Chunk, Triple, Triples, Metadata, Value
from .... schema import Chunk, Triple, Triples, Metadata, Term, IRI, LITERAL
from .... schema import chunk_ingest_queue, triples_store_queue
from .... schema import prompt_request_queue
from .... schema import prompt_response_queue
@ -20,7 +20,7 @@ from .... clients.prompt_client import PromptClient
from .... rdf import TRUSTGRAPH_ENTITIES, DEFINITION
from .... base import ConsumerProducer
DEFINITION_VALUE = Value(value=DEFINITION, is_uri=True)
DEFINITION_VALUE = Term(type=IRI, iri=DEFINITION)
module = "kg-extract-topics"
@ -106,8 +106,8 @@ class Processor(ConsumerProducer):
s_uri = self.to_uri(s)
s_value = Value(value=str(s_uri), is_uri=True)
o_value = Value(value=str(o), is_uri=False)
s_value = Term(type=IRI, iri=str(s_uri))
o_value = Term(type=LITERAL, value=str(o))
await self.emit_edge(
v.metadata, s_value, DEFINITION_VALUE, o_value