32 lines
1.3 KiB
Python
32 lines
1.3 KiB
Python
"""Runtime Evidence lookup bound to the atomically active corpus generation."""
|
|
|
|
from tht.corpus.store import CorpusStore
|
|
|
|
|
|
def active_evidence_hits(store: CorpusStore, vector_store, embedding, *, limit: int):
|
|
manifest = store.active_manifest()
|
|
if manifest is None or manifest.vector_generation is None:
|
|
return []
|
|
document_generations = dict(manifest.metadata.get("document_generations", {}))
|
|
active_documents = {document.document_id for document in manifest.documents}
|
|
hits = vector_store.search(["evidence"], embedding, limit=limit, kinds=["evidence"])
|
|
return [
|
|
hit for hit in hits
|
|
if hit.metadata.get("document_id") in active_documents
|
|
and hit.metadata.get("vector_generation")
|
|
== document_generations.get(hit.metadata.get("document_id"), manifest.vector_generation)
|
|
]
|
|
|
|
|
|
def resolve_evidence_file(store: CorpusStore, evidence_id: str) -> str:
|
|
manifest = store.active_manifest()
|
|
if manifest is None:
|
|
return ""
|
|
for document in manifest.documents:
|
|
frontmatter = document.metadata.get("frontmatter", {})
|
|
identifiers = {document.document_id, document.source_id, str(frontmatter.get("id", ""))}
|
|
if evidence_id in identifiers:
|
|
path = store.resolve_document(document.document_id)
|
|
return str(path) if path else ""
|
|
return ""
|