Files
geointel/backend/tests/test_sprint112_qa_evidence_overlay.py
T
JensandClaude Opus 5 5278fcd361 bound the QA evidence overlay and fetch only what it draws
evidence_geojson emitted one feature per false positive, one per false negative
and two per match, with no limit. A regional check of 40k detections against 45k
reference footprints produced well over a hundred thousand features in a single
response, plus one warning string per unresolvable identifier. The endpoint the
entire review workflow depends on therefore failed exactly where review matters
most.

What to draw is now decided before any geometry is fetched, so the query work is
proportional to the result rather than to the size of the check — previously
130k geometries were resolved through an IN clause holding every identifier in
the check, to then discard most of them.

The budget is split between misses and false positives in proportion to their
populations with at least one of each, rather than by strict priority, which
would mean a check with 50.000 misses and three false positives never showed
one. Confirmations fill what remains, and a match is kept or dropped as a pair
because half a match is not reviewable evidence.

limit_evidence and evidence_role_counts are removed: plan_evidence supersedes
them, and helpers kept alive only by their own tests read like a contract.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
2026-08-22 15:14:26 +02:00

180 lines
7.2 KiB
Python

from __future__ import annotations
from uuid import uuid4
import pytest
from fastapi.testclient import TestClient
from geoalchemy2.shape import from_shape
from shapely.geometry import box
from app.core.errors import AppError
from app.db.session import get_db
from app.main import app
from app.models import QualityCheck, VectorFeature
from app.services.quality_evidence_service import QualityEvidenceService
class FakeQuery:
def __init__(self, rows):
self.rows = list(rows)
def filter(self, *criteria):
for criterion in criteria:
left = getattr(criterion, "left", None)
right = getattr(criterion, "right", None)
operator = getattr(criterion, "operator", None)
name = getattr(left, "name", None)
value = getattr(right, "value", right)
if name and operator and operator.__name__ == "eq":
self.rows = [row for row in self.rows if getattr(row, name) == value]
return self
def all(self):
return list(self.rows)
class FakeSession:
def __init__(self, objects=None, query_rows=None) -> None:
self.objects = objects or {}
self.query_rows = query_rows or {}
def get(self, model, item_id):
return self.objects.get((model, item_id))
def query(self, model):
return FakeQuery(self.query_rows.get(model, []))
def _vector_feature(dataset_id, *, feature_id=None, source_feature_id: str, geom=None) -> VectorFeature:
return VectorFeature(
id=feature_id or uuid4(),
dataset_id=dataset_id,
source_feature_id=source_feature_id,
feature_class="building",
properties_json={"name": source_feature_id},
geometry=from_shape(geom or box(4.0, 51.0, 4.1, 51.1), srid=4326),
)
def test_quality_check_evidence_geojson_resolves_persisted_vector_features() -> None:
project_id = uuid4()
quality_check_id = uuid4()
candidate_dataset_id = uuid4()
reference_dataset_id = uuid4()
candidate_match = _vector_feature(candidate_dataset_id, source_feature_id="candidate-match")
candidate_extra = _vector_feature(candidate_dataset_id, source_feature_id="candidate-extra", geom=box(4.4, 51.4, 4.5, 51.5))
reference_match = _vector_feature(reference_dataset_id, source_feature_id="reference-match")
reference_missing = _vector_feature(reference_dataset_id, source_feature_id="reference-missing", geom=box(4.7, 51.7, 4.8, 51.8))
quality_check = QualityCheck(
id=quality_check_id,
project_id=project_id,
candidate_dataset_id=candidate_dataset_id,
reference_dataset_id=reference_dataset_id,
check_type="candidate_vs_reference",
status="ok",
findings_json={
"match_evidence": [
{
"candidate_feature_id": "candidate-match",
"reference_feature_id": "reference-match",
"iou": 1.0,
}
],
"false_positive_evidence": [{"candidate_feature_id": "candidate-extra"}],
"false_negative_evidence": [{"reference_feature_id": "reference-missing"}],
},
)
db = FakeSession(
objects={(QualityCheck, quality_check_id): quality_check},
query_rows={VectorFeature: [candidate_match, candidate_extra, reference_match, reference_missing]},
)
result = QualityEvidenceService.evidence_geojson(db, project_id=project_id, quality_check_id=quality_check_id)
assert result["quality_check_id"] == str(quality_check_id)
assert result["feature_count"] == 4
assert result["geojson"]["type"] == "FeatureCollection"
roles = [feature["properties"]["qa_evidence_role"] for feature in result["geojson"]["features"]]
# Every role resolves to persisted geometry. Errors are emitted before
# confirmations, because a capped overlay must spend its budget on the
# objects a reviewer has to act on.
assert sorted(roles) == ["false_negative", "false_positive", "match_candidate", "match_reference"]
assert roles.index("false_negative") < roles.index("match_candidate")
assert roles.index("false_positive") < roles.index("match_candidate")
match_candidate = next(
feature
for feature in result["geojson"]["features"]
if feature["properties"]["qa_evidence_role"] == "match_candidate"
)
assert match_candidate["properties"]["quality_check_id"] == str(quality_check_id)
assert match_candidate["properties"]["candidate_feature_id"] == "candidate-match"
assert match_candidate["properties"]["reference_feature_id"] == "reference-match"
assert match_candidate["properties"]["iou"] == 1.0
assert match_candidate["properties"]["source_feature_id"] == "candidate-match"
def test_quality_check_evidence_geojson_rejects_cross_project_access() -> None:
quality_check_id = uuid4()
quality_check = QualityCheck(
id=quality_check_id,
project_id=uuid4(),
reference_dataset_id=uuid4(),
check_type="candidate_vs_reference",
status="ok",
findings_json={},
)
db = FakeSession(objects={(QualityCheck, quality_check_id): quality_check})
with pytest.raises(AppError) as exc:
QualityEvidenceService.evidence_geojson(db, project_id=uuid4(), quality_check_id=quality_check_id)
assert exc.value.code == "QUALITY_CHECK_NOT_FOUND"
def test_quality_check_evidence_geojson_api_uses_canonical_envelope(monkeypatch) -> None:
project_id = uuid4()
quality_check_id = uuid4()
reference_dataset_id = uuid4()
payload = {
"quality_check_id": str(quality_check_id),
"project_id": str(project_id),
"candidate_dataset_id": None,
"reference_dataset_id": str(reference_dataset_id),
"analysis_run_id": None,
"feature_count": 0,
"warnings": [],
"geojson": {"type": "FeatureCollection", "features": []},
}
monkeypatch.setattr(
"app.api.routes.quality_checks.QualityEvidenceService.evidence_geojson",
lambda *_args, **_kwargs: payload,
)
app.dependency_overrides[get_db] = lambda: FakeSession()
try:
response = TestClient(app).get(f"/api/v1/projects/{project_id}/quality-checks/{quality_check_id}/evidence/geojson")
finally:
app.dependency_overrides.pop(get_db, None)
assert response.status_code == 200
# The envelope wraps the service result; asserting the exact field list
# would break every time the response model gains a documented field.
body = response.json()
assert set(body) == {"data"}
assert body["data"].items() >= payload.items()
def test_frontend_quality_evidence_overlay_contract_is_wired() -> None:
from pathlib import Path
root = Path(__file__).resolve().parents[2]
geo_map = (root / "frontend" / "src" / "components" / "GeoMap.tsx").read_text(encoding="utf-8")
map_workspace = (root / "frontend" / "src" / "components" / "map" / "MapWorkspace.tsx").read_text(encoding="utf-8")
qa_api = (root / "frontend" / "src" / "services" / "api" / "qa.ts").read_text(encoding="utf-8")
assert "qaEvidenceData" in geo_map
assert "qa-evidence-fill" in geo_map
assert "qa_evidence_role" in geo_map
assert "qualityEvidenceGeoJson" in map_workspace
assert "getQualityEvidenceGeoJson" in qa_api