Files

104 lines
3.3 KiB
Python
Raw Permalink Normal View History

2026-09-04 17:34:03 -05:00
"""Person ingestion pipeline (section 15).
Movie/series ingestion already creates a stub `Person` row (name + gender)
for each cast/crew credit via `get_or_create_person`. This module fetches
and upserts a *full* profile (biography, birthday, images, ...) for one
person, either standalone (admin-triggered) or as a follow-up job.
"""
from __future__ import annotations
import uuid
from sqlalchemy.ext.asyncio import AsyncSession
from app.core.logging import get_logger
from app.models.enums import MatchConfidence
2026-09-04 17:34:03 -05:00
from app.models.person import Person
from app.providers.base import MetadataProvider
from app.repositories import polymorphic
2026-09-04 17:34:03 -05:00
from app.repositories import reference_repository as refs
from app.repositories.provider_repository import get_or_create_provider, upsert_provider_record
from app.services.metadata.matcher import match_entity
from app.services.metadata.normalizer import normalize_person
logger = get_logger(__name__)
async def ingest_person(
session: AsyncSession,
provider: MetadataProvider,
external_id: str,
*,
ingestion_job_id: uuid.UUID | None = None,
) -> uuid.UUID:
tmdb_provider = await get_or_create_provider(session, provider.slug, provider.name)
aurora_id_existing, confidence = await match_entity(
session, provider_id=tmdb_provider.id, entity_type="person", external_id=external_id
)
raw = await provider.get_person(external_id)
normalized = normalize_person(raw)
person: Person | None = None
if aurora_id_existing is not None:
person = await session.get(Person, aurora_id_existing)
if person is None:
person = Person(name=normalized.name)
session.add(person)
await session.flush()
aurora_id_existing = None
person.name = normalized.name
person.biography = normalized.biography
person.birthday = normalized.birthday
person.deathday = normalized.deathday
person.birthplace = normalized.birthplace
person.gender = normalized.gender
person.known_for_department = normalized.known_for_department
person.popularity = normalized.popularity
person.aliases = normalized.aliases
await session.flush()
await polymorphic.sync_artwork(
session,
entity_type="person",
entity_id=person.id,
provider_id=tmdb_provider.id,
artwork_list=normalized.artwork,
2026-09-04 17:34:03 -05:00
)
await refs.link_external_id(
session,
entity_type="person",
entity_id=person.id,
provider_id=tmdb_provider.id,
external_id=external_id,
)
for slug, ext_id in normalized.external_ids.items():
ext_provider = await get_or_create_provider(session, slug, slug.upper())
await refs.link_external_id(
session,
entity_type="person",
entity_id=person.id,
provider_id=ext_provider.id,
external_id=ext_id,
)
await upsert_provider_record(
session,
provider_id=tmdb_provider.id,
entity_type="person",
external_id=external_id,
raw_data=raw,
aurora_entity_id=person.id,
match_confidence=confidence if aurora_id_existing else MatchConfidence.NEW,
ingestion_job_id=ingestion_job_id,
)
logger.info("person_ingested", person_id=str(person.id), external_id=external_id)
return person.id