Ingest and self-host company/network logos

CompanyBriefOut.logo_path and CompanyDetailOut.logo_path have existed
in the API schemas since Core 0.1's first pass, but nothing ever
actually populated them -- TMDb's production_companies/networks
entries carry a logo_path field that the normalizer simply never read,
so every response has silently returned null there.

- NormalizedCompanyRef now captures logo_path from both movie
  production_companies and series networks/production_companies.
- normalizer.company_logo_artwork() builds the single-image artwork
  entry for it (a company has exactly one logo ingested, so it's
  always is_primary) -- centralized there rather than duplicated in
  both ingestion modules, matching how movie/series artwork URLs are
  already built in one place.
- movie_ingestion.py and series_ingestion.py sync it via the same
  sync_artwork() the rest of section 10's self-hosting already uses,
  so it gets the same re-ingestion-safe upsert behavior for free.
- tasks.py cascades a download_artwork_task per referenced company
  after movie/series ingestion, same fan-out pattern as
  movie/series/season/person artwork.

Co-Authored-By: Claude Sonnet 5 <[email protected]>

app/services/ingestion/movie_ingestion.py

@@ -32,7 +32,7 @@ from app.repositories import polymorphic
 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 NormalizedMovie, normalize_movie
+from app.services.metadata.normalizer import NormalizedMovie, company_logo_artwork, normalize_movie
 
 logger = get_logger(__name__)
 
@@ -68,6 +68,15 @@ async def _replace_movie_companies(
             country_code=country_code,
         )
         session.add(MovieCompany(movie_id=movie_id, company_id=company.id, role=ref.role))
+        logo = company_logo_artwork(ref.logo_path)
+        if logo is not None:
+            await polymorphic.sync_artwork(
+                session,
+                entity_type="company",
+                entity_id=company.id,
+                provider_id=provider_id,
+                artwork_list=[logo],
+            )
 
 
 async def _replace_movie_credits(

app/services/ingestion/runner.py

@@ -19,6 +19,7 @@ from app.models.enums import IngestionJobStatus
 from app.models.provider import IngestionJob
 from app.providers.base import MetadataProvider
 from app.providers.tmdb import TMDbProvider
+from app.repositories.movie_repository import get_movie_companies
 from app.repositories.polymorphic import get_primary_artwork_ids
 from app.repositories.provider_repository import (
     create_ingestion_job,
@@ -28,7 +29,7 @@ from app.repositories.provider_repository import (
     mark_job_running,
     mark_job_succeeded,
 )
-from app.repositories.tv_repository import get_seasons_for_series
+from app.repositories.tv_repository import get_seasons_for_series, get_series_companies
 from app.services.ingestion.movie_ingestion import ingest_movie
 from app.services.ingestion.person_ingestion import ingest_person
 from app.services.ingestion.series_ingestion import ingest_season, ingest_series
@@ -185,3 +186,25 @@ async def get_primary_artwork_for_download(
 
     async with session_scope() as session:
         return await get_primary_artwork_ids(session, entity_type, entity_ids)
+
+
+async def get_movie_company_ids(movie_id: uuid.UUID) -> list[uuid.UUID]:
+    """Companies (production companies) a just-ingested movie references,
+    for cascading their logo into a download too — see
+    app/workers/tasks.py:_cascade_artwork_downloads. Shared across many
+    movies, so this fires on every movie referencing a given company, but
+    download_artwork is idempotent (skips once local_path is set)."""
+
+    async with session_scope() as session:
+        companies = await get_movie_companies(session, movie_id)
+        return [company.id for company, _role in companies]
+
+
+async def get_series_company_ids(series_id: uuid.UUID) -> list[uuid.UUID]:
+    """Same as get_movie_company_ids, but for a series' companies *and*
+    networks (get_series_companies with no company_type filter returns
+    both)."""
+
+    async with session_scope() as session:
+        companies = await get_series_companies(session, series_id)
+        return [company.id for company, _role in companies]

app/services/ingestion/series_ingestion.py

@@ -34,6 +34,7 @@ from app.repositories.provider_repository import get_or_create_provider, upsert_
 from app.services.metadata.matcher import match_entity
 from app.services.metadata.normalizer import (
     NormalizedSeries,
+    company_logo_artwork,
     normalize_season,
     normalize_series,
 )
@@ -94,6 +95,15 @@ async def _replace_series_companies(
             country_code=country_code,
         )
         session.add(SeriesCompany(series_id=series_id, company_id=company.id, role=role))
+        logo = company_logo_artwork(ref.logo_path)
+        if logo is not None:
+            await polymorphic.sync_artwork(
+                session,
+                entity_type="company",
+                entity_id=company.id,
+                provider_id=provider_id,
+                artwork_list=[logo],
+            )
 
 
 async def _replace_series_credits(

app/services/metadata/normalizer.py

@@ -54,6 +54,7 @@ class NormalizedCompanyRef:
     name: str
     country_code: str | None = None
     role: str = "production"
+    logo_path: str | None = None  # TMDb's raw path, e.g. "/logo.png" — not a full URL yet
 
 
 @dataclass
@@ -74,6 +75,27 @@ class NormalizedArtwork:
     is_primary: bool = False
 
 
+def company_logo_artwork(logo_path: str | None) -> NormalizedArtwork | None:
+    """A single-item artwork entry for a company/network's TMDb logo_path,
+    or None if it doesn't have one. Shared by movie_ingestion.py and
+    series_ingestion.py so the "how do we turn a relative TMDb path into a
+    real image URL" logic stays in one place (this module), rather than the
+    ingestion layer needing to know about TMDb's CDN itself."""
+
+    if not logo_path:
+        return None
+    return NormalizedArtwork(
+        artwork_type="logo",
+        url=_image_url_required(logo_path),
+        provider_asset_id=logo_path,
+        width=None,
+        height=None,
+        aspect_ratio=None,
+        language=None,
+        is_primary=True,  # a company has exactly one logo ingested today
+    )
+
+
 @dataclass
 class NormalizedVideo:
     video_type: str
@@ -173,6 +195,7 @@ def normalize_movie(raw: dict) -> NormalizedMovie:
             external_id=str(pc["id"]),
             name=pc["name"],
             country_code=pc.get("origin_country") or None,
+            logo_path=pc.get("logo_path"),
         )
         for pc in raw.get("production_companies", [])
     ]
@@ -464,7 +487,10 @@ def normalize_series(raw: dict) -> NormalizedSeries:
 
     networks = [
         NormalizedCompanyRef(
-            external_id=str(n["id"]), name=n["name"], country_code=n.get("origin_country") or None
+            external_id=str(n["id"]),
+            name=n["name"],
+            country_code=n.get("origin_country") or None,
+            logo_path=n.get("logo_path"),
         )
         for n in raw.get("networks", [])
     ]
@@ -473,6 +499,7 @@ def normalize_series(raw: dict) -> NormalizedSeries:
             external_id=str(pc["id"]),
             name=pc["name"],
             country_code=pc.get("origin_country") or None,
+            logo_path=pc.get("logo_path"),
         )
         for pc in raw.get("production_companies", [])
     ]

app/workers/tasks.py

@@ -22,8 +22,10 @@ from app.core.logging import get_logger
 from app.services.artwork.download import download_artwork
 from app.services.artwork.storage import get_artwork_storage
 from app.services.ingestion.runner import (
+    get_movie_company_ids,
     get_or_create_ingestion_job,
     get_primary_artwork_for_download,
+    get_series_company_ids,
     get_series_season_ids,
     get_series_season_numbers,
     run_movie_ingestion,
@@ -52,6 +54,7 @@ async def _cascade_artwork_downloads(
 async def ingest_movie_task(ctx: dict[str, Any], job_id: str, external_id: str) -> str:
     movie_id = await run_movie_ingestion(uuid.UUID(job_id), external_id)
     await _cascade_artwork_downloads(ctx, "movie", [movie_id])
+    await _cascade_artwork_downloads(ctx, "company", await get_movie_company_ids(movie_id))
     return str(movie_id)
 
 
@@ -70,6 +73,7 @@ async def ingest_series_task(
     # creates, independent of include_seasons (which only controls whether
     # full per-season episode data gets fetched) -- so this always runs.
     await _cascade_artwork_downloads(ctx, "season", await get_series_season_ids(series_id))
+    await _cascade_artwork_downloads(ctx, "company", await get_series_company_ids(series_id))
 
     if include_seasons:
         # Fan out one ingest_season job per season this series has, using

docs/database.md

@@ -82,8 +82,11 @@ automatically once it has.
 
 Only artwork marked `is_primary` gets downloaded automatically — the single
 image the provider designates as *the* image for an entity (TMDb's
-`poster_path`/`backdrop_path`/`profile_path`), not the dozens of alternates
-also stored. `app/services/ingestion/runner.py:get_primary_artwork_for_download`
+`poster_path`/`backdrop_path`/`profile_path`/`logo_path`), not the dozens of
+alternates also stored. This includes company/network logos (a company has
+exactly one ingested, so it's always primary) — cascaded per-movie/series
+since companies aren't independently ingested the way movies/series/people
+are. `app/services/ingestion/runner.py:get_primary_artwork_for_download`
 finds them and `app/workers/tasks.py` enqueues a `download_artwork_task` per
 id once movie/series/person/season ingestion succeeds — see
 `docs/architecture.md` for the full flow.

tests/integration/test_movie_ingestion.py

@@ -11,6 +11,8 @@ import pytest
 from sqlalchemy import select
 from sqlalchemy.ext.asyncio import AsyncSession
 
+from app.models.artwork import Artwork
+from app.models.company import Company
 from app.models.credit import Credit
 from app.models.genre import Genre
 from app.models.movie import Movie
@@ -103,3 +105,26 @@ async def test_ingest_movie_is_idempotent(db_session: AsyncSession):
     )
     # Credits are replaced wholesale on re-sync, not duplicated.
     assert len(credits_) == 2  # 1 cast + 1 crew from FAKE_MOVIE
+
+
+@pytest.mark.asyncio
+async def test_ingest_movie_ingests_company_logo(db_session: AsyncSession):
+    provider = FakeTMDbProvider(FAKE_MOVIE)
+
+    await ingest_movie(db_session, provider, "603")
+    await db_session.commit()
+
+    company = (
+        await db_session.execute(
+            select(Company).where(Company.name == "Village Roadshow Pictures")
+        )
+    ).scalar_one()
+
+    logo = (
+        await db_session.execute(
+            select(Artwork).where(Artwork.entity_type == "company", Artwork.entity_id == company.id)
+        )
+    ).scalar_one()
+    assert logo.artwork_type.value == "logo"
+    assert logo.is_primary is True
+    assert logo.original_url.endswith("/village-roadshow-logo.png")

tests/unit/test_normalizer.py

@@ -32,7 +32,12 @@ FAKE_MOVIE = {
     "imdb_id": "tt0133093",
     "genres": [{"id": 28, "name": "Action"}, {"id": 878, "name": "Science Fiction"}],
     "production_companies": [
-        {"id": 79, "name": "Village Roadshow Pictures", "origin_country": "US"}
+        {
+            "id": 79,
+            "name": "Village Roadshow Pictures",
+            "origin_country": "US",
+            "logo_path": "/village-roadshow-logo.png",
+        }
     ],
     "poster_path": "/poster.jpg",
     "backdrop_path": "/backdrop.jpg",
@@ -160,6 +165,17 @@ def test_normalize_movie_marks_exactly_one_primary_poster_and_backdrop():
     assert primary_backdrops[0].provider_asset_id == "/backdrop.jpg"
 
 
+def test_normalize_movie_maps_company_logo_path():
+    movie = normalize_movie(FAKE_MOVIE)
+    assert movie.companies[0].logo_path == "/village-roadshow-logo.png"
+
+
+def test_normalize_movie_company_without_logo_is_none():
+    minimal = {**FAKE_MOVIE, "production_companies": [{"id": 1, "name": "No Logo Studio"}]}
+    movie = normalize_movie(minimal)
+    assert movie.companies[0].logo_path is None
+
+
 def test_normalize_movie_external_ids_never_include_tmdb_itself():
     # TMDb's own id is handled separately by the ingestion pipeline via
     # `normalized.external_id`, not folded into `external_ids` (section 3:
@@ -224,9 +240,9 @@ FAKE_SERIES = {
     "vote_count": 15000,
     "genres": [{"id": 18, "name": "Drama"}, {"id": 80, "name": "Crime"}],
     "created_by": [{"id": 66633, "name": "Vince Gilligan", "gender": 2}],
-    "networks": [{"id": 174, "name": "AMC", "origin_country": "US"}],
+    "networks": [{"id": 174, "name": "AMC", "origin_country": "US", "logo_path": "/amc-logo.png"}],
     "production_companies": [
-        {"id": 2605, "name": "Gran Via Productions", "origin_country": "US"}
+        {"id": 2605, "name": "Gran Via Productions", "origin_country": "US", "logo_path": None}
     ],
     "poster_path": "/poster.jpg",
     "backdrop_path": "/backdrop.jpg",
@@ -323,6 +339,14 @@ def test_normalize_series_maps_networks_vs_production_companies():
     assert [c.name for c in series.companies] == ["Gran Via Productions"]
 
 
+def test_normalize_series_maps_network_logo_path():
+    series = normalize_series(FAKE_SERIES)
+    assert series.networks[0].logo_path == "/amc-logo.png"
+    # Gran Via's fixture entry has no logo_path at all -- must come through
+    # as None, not KeyError.
+    assert series.companies[0].logo_path is None
+
+
 def test_normalize_series_maps_creators_separately_from_crew():
     series = normalize_series(FAKE_SERIES)
     assert len(series.creators) == 1