test: enforce 80% pytest coverage gate across all 6 Python domains (#441)

.cspell/general-technical.txt

@@ -2035,3 +2035,20 @@ vsmserver
 envaccount
 envcontainer
 mycontainer
+anoms
+caplog
+cfgs
+ckpts
+coro
+dotdot
+errstate
+excinfo
+fromlist
+ftypmp
+jpegbytes
+matroska
+myjob
+reraises
+topcam
+notanum
+fevil

.github/workflows/dataviewer-backend-pytests.yml

@@ -43,7 +43,7 @@ jobs:
         run: uv sync --extra dev --extra analysis --extra hdf5 --extra export --extra auth
 
       - name: Run pytest with coverage
-        run: uv run pytest -v --cov=src --cov-report=xml:../../../logs/coverage-dataviewer.xml --cov-report=term-missing
+        run: uv run pytest -v --cov=src --cov-report=xml:../../../logs/coverage-dataviewer.xml --cov-report=term-missing --cov-fail-under=90
 
       - name: Upload coverage.xml artifact
         if: ${{ inputs.code-coverage && always() }}

.github/workflows/main.yml

@@ -157,6 +157,16 @@ jobs:
       contents: read
       id-token: write
 
+  # Evaluation domain pytest execution
+  evaluation-pytests:
+    name: Evaluation Pytest
+    uses: ./.github/workflows/evaluation-pytests.yml
+    with:
+      code-coverage: true
+    permissions:
+      contents: read
+      id-token: write
+
   # Fuzz regression via deterministic corpus-based tests
   fuzz-regression-tests:
     name: Fuzz Regression Tests
@@ -268,6 +278,7 @@ jobs:
       - pytest-inference
       - dataviewer-frontend-tests
       - dataviewer-backend-pytests
+      - evaluation-pytests
       - python-lint
       - terraform-lint
       - terraform-validation

.github/workflows/pytest-data-pipeline.yml

@@ -43,6 +43,7 @@ jobs:
           --cov=data-pipeline/capture
           --cov-report=term-missing
           --cov-report=xml:logs/coverage-data-pipeline.xml
+          --cov-fail-under=80
 
       - name: Upload coverage.xml artifact
         if: always()

.github/workflows/pytest-dm-tools.yml

@@ -43,6 +43,7 @@ jobs:
           --cov=data-management/tools
           --cov-report=term-missing
           --cov-report=xml:logs/coverage-dm-tools.xml
+          --cov-fail-under=80
 
       - name: Upload coverage.xml artifact
         if: always()

.github/workflows/pytest-inference.yml

@@ -43,6 +43,7 @@ jobs:
           --cov=fleet-deployment/inference
           --cov-report=term-missing
           --cov-report=xml:logs/coverage-inference.xml
+          --cov-fail-under=80
 
       - name: Upload coverage.xml artifact
         if: always()

.github/workflows/pytest-training.yml

@@ -43,6 +43,7 @@ jobs:
           --cov=training
           --cov-report=term-missing
           --cov-report=xml:logs/coverage-training.xml
+          --cov-fail-under=80
 
       - name: Upload coverage.xml artifact
         if: always()

data-management/viewer/backend/pyproject.toml

@@ -78,3 +78,12 @@ constraint-dependencies = ["pygments==2.20.0"]
 [tool.pytest.ini_options]
 asyncio_mode = "auto"
 testpaths = ["tests"]
+
+[tool.coverage.run]
+source = ["src"]
+branch = true
+omit = ["src/**/conftest.py", "src/**/__init__.py"]
+
+[tool.coverage.report]
+show_missing = true
+precision = 2

data-management/viewer/backend/src/api/routers/labels.py

@@ -31,6 +31,11 @@
 if TYPE_CHECKING:
     from ..storage.blob_dataset import BlobDatasetProvider
 
+try:
+    from azure.storage.blob import ContentSettings
+except ImportError:
+    ContentSettings = None
+
 logger = logging.getLogger(__name__)
 
 router = APIRouter()
@@ -142,16 +147,15 @@ async def load(self, dataset_id: str) -> DatasetLabelsFile:
 
     async def save(self, dataset_id: str, labels_file: DatasetLabelsFile) -> None:
         try:
-            from azure.storage.blob import ContentSettings
-
             client = await self._provider._get_client()
             container = client.get_container_client(self._provider.container_name)
             blob_client = container.get_blob_client(self._blob_path(dataset_id))
             content = json.dumps(labels_file.model_dump(), indent=2).encode("utf-8")
+            content_settings = ContentSettings(content_type="application/json") if ContentSettings is not None else None
             await blob_client.upload_blob(
                 content,
                 overwrite=True,
-                content_settings=ContentSettings(content_type="application/json"),
+                content_settings=content_settings,
             )
         except Exception as e:
             logger.error(

data-management/viewer/backend/src/api/services/dataset_service/service.py

@@ -652,11 +652,13 @@ async def _prefetch() -> None:
         # Clean up completed tasks
         self._prefetch_tasks = {t for t in self._prefetch_tasks if not t.done()}
 
+        coro = _prefetch()
         try:
-            task = asyncio.create_task(_prefetch())
+            task = asyncio.create_task(coro)
             self._prefetch_tasks.add(task)
             task.add_done_callback(self._prefetch_tasks.discard)
         except RuntimeError as error:
+            coro.close()
             logger.debug("Skipping episode prefetch for episode %d: %s", int(episode_idx), error)
 
     def is_safe_video_path(self, video_path: str) -> bool:

data-management/viewer/backend/src/api/storage/azure.py

@@ -22,8 +22,17 @@
 
     AZURE_AVAILABLE = True
 except ImportError:
-    HttpResponseError = Exception
-    ResourceNotFoundError = Exception
+    # Distinct sentinel subclasses so `except` clauses don't accidentally
+    # match unrelated exceptions when the SDK isn't installed. Tests patch
+    # these module attributes to inject their own classes.
+    class _HttpResponseErrorStub(Exception):
+        """Sentinel for HttpResponseError when azure SDK is unavailable."""
+
+    class _ResourceNotFoundErrorStub(Exception):
+        """Sentinel for ResourceNotFoundError when azure SDK is unavailable."""
+
+    HttpResponseError = _HttpResponseErrorStub
+    ResourceNotFoundError = _ResourceNotFoundErrorStub
     DefaultAzureCredential = None
     ContentSettings = None
     BlobServiceClient = None

data-management/viewer/backend/tests/api/models/test_detection.py

@@ -0,0 +1,58 @@
+"""Tests for detection Pydantic models."""
+
+from __future__ import annotations
+
+import pytest
+from pydantic import ValidationError
+
+from src.api.models.detection import (
+    ClassSummary,
+    Detection,
+    DetectionRequest,
+    DetectionResult,
+    EpisodeDetectionSummary,
+)
+
+
+class TestDetectionRequest:
+    def test_defaults(self):
+        req = DetectionRequest()
+        assert req.frames is None
+        assert req.confidence == 0.25
+        assert req.model == "yolo11n"
+
+    def test_validate_frames_none_returns_none(self):
+        req = DetectionRequest(frames=None)
+        assert req.frames is None
+
+    def test_validate_frames_valid_list(self):
+        req = DetectionRequest(frames=[0, 1, 5, 10])
+        assert req.frames == [0, 1, 5, 10]
+
+    def test_validate_frames_negative_raises(self):
+        with pytest.raises(ValidationError, match="non-negative"):
+            DetectionRequest(frames=[0, -1, 2])
+
+    def test_confidence_out_of_range(self):
+        with pytest.raises(ValidationError):
+            DetectionRequest(confidence=1.5)
+
+
+class TestDetectionModels:
+    def test_detection_instantiation(self):
+        det = Detection(class_id=0, class_name="person", confidence=0.9, bbox=(0.0, 0.0, 10.0, 20.0))
+        assert det.class_id == 0
+        assert det.bbox == (0.0, 0.0, 10.0, 20.0)
+
+    def test_detection_result_defaults(self):
+        result = DetectionResult(frame=3, processing_time_ms=12.5)
+        assert result.detections == []
+
+    def test_class_summary(self):
+        summary = ClassSummary(count=4, avg_confidence=0.75)
+        assert summary.count == 4
+
+    def test_episode_summary_defaults(self):
+        summary = EpisodeDetectionSummary(total_frames=10, processed_frames=5, total_detections=2)
+        assert summary.detections_by_frame == []
+        assert summary.class_summary == {}