feat(provenance): _artists_for resolves via artist_id (closes vii-b gap)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-05-18 21:36:16 -04:00
parent aaf5d1ea70
commit e798302cfa
2 changed files with 27 additions and 77 deletions
+9 -34
View File
@@ -13,7 +13,7 @@ from datetime import datetime
from sqlalchemy import and_, exists, func, or_, select
from sqlalchemy.ext.asyncio import AsyncSession
from ..models import Artist, ImageProvenance, ImageRecord, Post, Source, Tag
from ..models import Artist, ImageProvenance, ImageRecord, Source, Tag
from ..models.tag import image_tag
CURSOR_SEPARATOR = "|"
@@ -94,44 +94,19 @@ def _provenance_clause(post_id, artist_id):
async def _artists_for(session, image_ids: list[int]) -> dict[int, dict]:
"""Map image_id -> {"name","slug"} using the primary-post path first,
then the lowest-id ImageProvenance row as fallback. Bounded by page
size (<= 200) → two indexed selects, no N+1."""
"""Map image_id -> {"name","slug"} via the canonical
image_record.artist_id (FC-2d-vii-c). Bounded by page size."""
if not image_ids:
return {}
out: dict[int, dict] = {}
primary = (
stmt = (
select(ImageRecord.id, Artist.name, Artist.slug)
.join(Post, Post.id == ImageRecord.primary_post_id)
.join(Source, Source.id == Post.source_id)
.join(Artist, Artist.id == Source.artist_id)
.join(Artist, Artist.id == ImageRecord.artist_id)
.where(ImageRecord.id.in_(image_ids))
)
for img_id, name, slug in (await session.execute(primary)).all():
out[img_id] = {"name": name, "slug": slug}
remaining = [i for i in image_ids if i not in out]
if remaining:
fb = (
select(
ImageProvenance.image_record_id,
ImageProvenance.id,
Artist.name,
Artist.slug,
)
.join(Source, Source.id == ImageProvenance.source_id)
.join(Artist, Artist.id == Source.artist_id)
.where(ImageProvenance.image_record_id.in_(remaining))
.order_by(
ImageProvenance.image_record_id.asc(),
ImageProvenance.id.asc(),
)
)
for img_id, _pid, name, slug in (await session.execute(fb)).all():
if img_id not in out: # first (lowest id) wins
out[img_id] = {"name": name, "slug": slug}
return out
return {
img_id: {"name": name, "slug": slug}
for img_id, name, slug in (await session.execute(stmt)).all()
}
class GalleryService:
+18 -43
View File
@@ -3,13 +3,7 @@ from datetime import UTC, datetime, timedelta
import pytest
from backend.app import create_app
from backend.app.models import (
Artist,
ImageProvenance,
ImageRecord,
Post,
Source,
)
from backend.app.models import Artist, ImageRecord, Source
pytestmark = pytest.mark.integration
@@ -48,42 +42,6 @@ async def _artist_source(db, name, slug):
return a, s
@pytest.mark.asyncio
async def test_scroll_artist_via_primary_post(client, db):
rec = await _img(db, 1)
_, s = await _artist_source(db, "Alice", "alice")
post = Post(source_id=s.id, external_post_id="1")
db.add(post)
await db.flush()
rec.primary_post_id = post.id
await db.flush()
await db.commit()
resp = await client.get("/api/gallery/scroll?limit=10")
assert resp.status_code == 200
body = await resp.get_json()
item = next(i for i in body["images"] if i["id"] == rec.id)
assert item["artist"] == {"name": "Alice", "slug": "alice"}
@pytest.mark.asyncio
async def test_scroll_artist_via_provenance_fallback(client, db):
rec = await _img(db, 1)
_, s = await _artist_source(db, "Bob", "bob")
post = Post(source_id=s.id, external_post_id="2")
db.add(post)
await db.flush()
db.add(ImageProvenance(image_record_id=rec.id, post_id=post.id,
source_id=s.id))
await db.flush()
await db.commit() # rec.primary_post_id stays NULL
resp = await client.get("/api/gallery/scroll?limit=10")
body = await resp.get_json()
item = next(i for i in body["images"] if i["id"] == rec.id)
assert item["artist"] == {"name": "Bob", "slug": "bob"}
@pytest.mark.asyncio
async def test_scroll_artist_null_when_none(client, db):
rec = await _img(db, 1)
@@ -92,3 +50,20 @@ async def test_scroll_artist_null_when_none(client, db):
body = await resp.get_json()
item = next(i for i in body["images"] if i["id"] == rec.id)
assert item["artist"] is None
@pytest.mark.asyncio
async def test_scroll_artist_via_artist_id(client, db):
# FC-2d-vii-c: the canonical resolution path is image_record.artist_id
# (covers folder-style images with no provenance too).
rec = await _img(db, 1)
a, _ = await _artist_source(db, "Folder", "folder")
rec.artist_id = a.id
await db.flush()
await db.commit()
resp = await client.get("/api/gallery/scroll?limit=10")
assert resp.status_code == 200
body = await resp.get_json()
item = next(i for i in body["images"] if i["id"] == rec.id)
assert item["artist"] == {"name": "Folder", "slug": "folder"}