Initial public release

backend/tests/test_qa_service.py

@@ -0,0 +1,205 @@
+from __future__ import annotations
+
+import json
+from hashlib import sha256
+from pathlib import Path
+from uuid import uuid4
+
+from app.models import Dataset, SourceRegistry, SourceSnapshot
+from app.services.qa_service import QaService
+
+
+class FakeSession:
+    def __init__(self, datasets=None, areas=None):
+        self.datasets = {item.id: item for item in (datasets or [])}
+        self.areas = {item.id: item for item in (areas or [])}
+
+    def get(self, model, item_id):
+        if model.__name__ == "Dataset":
+            return self.datasets.get(item_id)
+        if model.__name__ == "Area":
+            return self.areas.get(item_id)
+        return None
+
+
+def _feature(feature_id: str, coordinates: list[list[list[float]]]) -> dict:
+    return {
+        "type": "Feature",
+        "id": feature_id,
+        "properties": {"source_feature_id": feature_id},
+        "geometry": {
+            "type": "Polygon",
+            "coordinates": coordinates,
+        },
+    }
+
+
+def _write_dataset(path: Path, coordinates: list[list[list[float]]]) -> None:
+    payload = {
+        "type": "FeatureCollection",
+        "features": [_feature("feature-1", coordinates)],
+    }
+    path.write_text(json.dumps(payload), encoding="utf-8")
+
+
+def _write_features(path: Path, features: list[dict]) -> None:
+    path.write_text(json.dumps({"type": "FeatureCollection", "features": features}), encoding="utf-8")
+
+
+def _authoritative_reference(dataset: Dataset) -> Dataset:
+    """Give QA reference fixtures the same durable authority proof as GRB."""
+
+    source_id = uuid4()
+    snapshot_id = uuid4()
+    checksum = sha256(Path(str(dataset.storage_path)).read_bytes()).hexdigest()
+    source = SourceRegistry(
+        id=source_id,
+        source_key="grb",
+        display_name="GRB test reference",
+        classification="authoritative",
+        authority_name="Digitaal Vlaanderen",
+        authority_scope_json={"zone": "Flanders"},
+        usage_policy_json={"ground_truth_allowed": True, "validation_authority": {"building_validation": "primary"}},
+    )
+    snapshot = SourceSnapshot(
+        id=snapshot_id,
+        source_registry_id=source_id,
+        snapshot_key=f"qa-grb-{dataset.id}",
+        checksum_sha256=checksum,
+        ingest_status="ingested",
+        freshness_status="current",
+    )
+    dataset.source = "grb"
+    dataset.source_name = "grb"
+    dataset.dataset_role = "reference"
+    dataset.status = "ready"
+    dataset.checksum_sha256 = checksum
+    dataset.source_registry_id = source_id
+    dataset.source_snapshot_id = snapshot_id
+    dataset.data_contract_key = "geointel.vector.geojson"
+    dataset.data_contract_version = "1.0.0"
+    dataset.validation_status = "passed"
+    dataset.provenance_status = "complete"
+    dataset.lineage_status = "complete"
+    dataset.quarantine_status = "not_quarantined"
+    dataset.source_registry = source
+    dataset.source_snapshot = snapshot
+    return dataset
+
+
+def test_qa_compare_candidate_with_reference_returns_metrics(tmp_path) -> None:
+    project_id = uuid4()
+    candidate_id = uuid4()
+    reference_id = uuid4()
+    candidate_path = tmp_path / "candidate.geojson"
+    reference_path = tmp_path / "reference.geojson"
+    polygon = [[[4.0, 51.0], [4.1, 51.0], [4.1, 51.1], [4.0, 51.1], [4.0, 51.0]]]
+    _write_dataset(candidate_path, polygon)
+    _write_dataset(reference_path, polygon)
+
+    candidate = Dataset(
+        id=candidate_id,
+        project_id=project_id,
+        name="candidate.geojson",
+        dataset_type="vector",
+        source="test",
+        storage_path=str(candidate_path),
+        crs="EPSG:4326",
+        metadata_json={"crs_assumed": False},
+    )
+    reference = _authoritative_reference(Dataset(
+        id=reference_id,
+        project_id=project_id,
+        name="reference.geojson",
+        dataset_type="vector",
+        source="test",
+        storage_path=str(reference_path),
+        crs="EPSG:4326",
+        metadata_json={"crs_assumed": False},
+    ))
+
+    result = QaService.compare_candidate_with_reference(
+        db=FakeSession([candidate, reference]),
+        project_id=project_id,
+        candidate_dataset_id=candidate_id,
+        reference_dataset_id=reference_id,
+        iou_threshold=0.5,
+    )
+
+    assert result.status == "ok"
+    assert result.matches == 1
+    assert result.false_positives == 0
+    assert result.false_negatives == 0
+    assert result.precision == 1.0
+    assert result.recall == 1.0
+    assert result.f1_score == 1.0
+
+
+def test_qa_compare_candidate_with_reference_returns_feature_level_evidence(tmp_path) -> None:
+    project_id = uuid4()
+    candidate_id = uuid4()
+    reference_id = uuid4()
+    candidate_path = tmp_path / "candidate.geojson"
+    reference_path = tmp_path / "reference.geojson"
+    matched_candidate = [[[4.0, 51.0], [4.1, 51.0], [4.1, 51.1], [4.0, 51.1], [4.0, 51.0]]]
+    matched_reference = [[[4.0, 51.0], [4.1, 51.0], [4.1, 51.1], [4.0, 51.1], [4.0, 51.0]]]
+    false_positive = [[[4.5, 51.5], [4.6, 51.5], [4.6, 51.6], [4.5, 51.6], [4.5, 51.5]]]
+    false_negative = [[[4.8, 51.8], [4.9, 51.8], [4.9, 51.9], [4.8, 51.9], [4.8, 51.8]]]
+    _write_features(candidate_path, [_feature("candidate-match", matched_candidate), _feature("candidate-extra", false_positive)])
+    _write_features(reference_path, [_feature("reference-match", matched_reference), _feature("reference-missing", false_negative)])
+
+    candidate = Dataset(
+        id=candidate_id,
+        project_id=project_id,
+        name="candidate.geojson",
+        dataset_type="vector",
+        source="test",
+        storage_path=str(candidate_path),
+        crs="EPSG:4326",
+        metadata_json={"crs_assumed": False},
+    )
+    reference = _authoritative_reference(Dataset(
+        id=reference_id,
+        project_id=project_id,
+        name="reference.geojson",
+        dataset_type="vector",
+        source="test",
+        storage_path=str(reference_path),
+        crs="EPSG:4326",
+        metadata_json={"crs_assumed": False},
+    ))
+
+    result = QaService.compare_candidate_with_reference(
+        db=FakeSession([candidate, reference]),
+        project_id=project_id,
+        candidate_dataset_id=candidate_id,
+        reference_dataset_id=reference_id,
+        iou_threshold=0.5,
+    )
+
+    assert result.matches == 1
+    assert result.false_positives == 1
+    assert result.false_negatives == 1
+    assert result.match_evidence == [
+        {
+            "candidate_feature_id": "candidate-match",
+            "reference_feature_id": "reference-match",
+            "iou": 1.0,
+        }
+    ]
+    assert result.false_positive_evidence == [{"candidate_feature_id": "candidate-extra"}]
+    assert result.false_negative_evidence == [{"reference_feature_id": "reference-missing"}]
+
+
+def test_dataset_reference_metadata_migration_declares_required_columns() -> None:
+    migration_path = Path(__file__).parents[1] / "alembic" / "versions" / "202606120001_add_dataset_reference_metadata.py"
+    migration_text = migration_path.read_text(encoding="utf-8")
+    for column_name in (
+        "dataset_role",
+        "source_name",
+        "reference_layer_name",
+        "source_metadata",
+        "provenance_metadata",
+        "imported_at",
+    ):
+        assert column_name in migration_text