Move tests to template and remove sample duplicates (#164)

Author: arthurhenrique

{{cookiecutter.project_slug}}/tests/__init__.py

@@ -0,0 +1,4 @@
+import os
+import sys
+
+sys.path.append(os.path.join(os.path.dirname(__file__), "..", "app"))

{{cookiecutter.project_slug}}/tests/test_api_predictor.py

@@ -0,0 +1,65 @@
+import json
+import pytest
+from fastapi.testclient import TestClient
+
+from main import get_application
+import api.routes.predictor as predictor
+from core import config as app_config
+import main as app_main
+
+
+@pytest.fixture
+def client(monkeypatch):
+    monkeypatch.setattr(app_config, "MEMOIZATION_FLAG", False)
+    monkeypatch.setattr(app_main, "MEMOIZATION_FLAG", False)
+    app = get_application()
+    return TestClient(app)
+
+
+@pytest.fixture
+def anyio_backend():
+    return "asyncio"
+
+
+def sample_payload():
+    return {
+        "feature1": 1.0,
+        "feature2": 2.0,
+        "feature3": 3.0,
+        "feature4": 4.0,
+        "feature5": 5.0,
+    }
+
+
+@pytest.mark.anyio
+async def test_predict_endpoint_success(monkeypatch):
+    monkeypatch.setattr(predictor, "get_prediction", lambda data: [1])
+    data = predictor.MachineLearningDataInput(**sample_payload())
+    resp = await predictor.predict(data)
+    assert resp.prediction == 1.0
+    assert resp.prediction_label == "label ok"
+
+
+def test_predict_endpoint_exception(client, monkeypatch):
+    def raise_error(data):
+        raise ValueError("fail")
+
+    monkeypatch.setattr(predictor, "get_prediction", raise_error)
+    response = client.post("/api/v1/predict", json=sample_payload())
+    assert response.status_code == 500
+
+
+def test_health_endpoint_success(client, monkeypatch, tmp_path):
+    example = tmp_path / "example.json"
+    example.write_text(json.dumps(sample_payload()))
+    monkeypatch.setattr(predictor, "INPUT_EXAMPLE", str(example))
+    monkeypatch.setattr(predictor, "get_prediction", lambda data: [0])
+    response = client.get("/api/v1/health")
+    assert response.status_code == 200
+    assert response.json() == {"status": True}
+
+
+def test_health_endpoint_failure(client, monkeypatch):
+    monkeypatch.setattr(predictor, "INPUT_EXAMPLE", "missing.json")
+    response = client.get("/api/v1/health")
+    assert response.status_code == 404

{{cookiecutter.project_slug}}/tests/test_config_and_errors.py

@@ -0,0 +1,18 @@
+import logging
+
+import pytest
+
+from core import config, errors
+
+
+def test_config_defaults():
+    assert config.API_PREFIX == "/api"
+    assert config.PROJECT_NAME == "{{cookiecutter.project_name}}"
+    assert config.LOGGING_LEVEL in (logging.INFO, logging.DEBUG)
+
+
+def test_custom_exceptions():
+    with pytest.raises(errors.PredictException):
+        raise errors.PredictException("test")
+    with pytest.raises(errors.ModelLoadException):
+        raise errors.ModelLoadException("test")

{{cookiecutter.project_slug}}/tests/test_events_and_main.py

@@ -0,0 +1,38 @@
+from fastapi import FastAPI
+
+from core import events
+from main import get_application
+import services.predict as predict
+
+
+def test_preload_model(monkeypatch):
+    called = {}
+
+    def fake_get_model(cls, loader):
+        called["called"] = True
+
+    monkeypatch.setattr(
+        predict.MachineLearningModelHandlerScore,
+        "get_model",
+        classmethod(fake_get_model),
+    )
+    events.preload_model()
+    assert called.get("called") is True
+
+
+def test_create_start_app_handler(monkeypatch):
+    called = {}
+
+    def fake_preload():
+        called["called"] = True
+
+    monkeypatch.setattr(events, "preload_model", fake_preload)
+    app = FastAPI()
+    handler = events.create_start_app_handler(app)
+    handler()
+    assert called.get("called") is True
+
+
+def test_get_application():
+    app = get_application()
+    assert isinstance(app, FastAPI)

{{cookiecutter.project_slug}}/tests/test_predict_service.py

@@ -0,0 +1,74 @@
+import os
+
+import pytest
+
+import services.predict as predict
+
+
+class DummyModel:
+    def predict(self, data):
+        return [42]
+
+
+class DummyScaler:
+    def transform(self, data):
+        return data
+
+
+def test_predict_success(monkeypatch):
+    predict.MachineLearningModelHandlerScore.model = DummyModel()
+    result = predict.MachineLearningModelHandlerScore.predict([[1]])
+    assert result == [42]
+
+
+def test_predict_missing_method(monkeypatch):
+    predict.MachineLearningModelHandlerScore.model = {"model": object(), "scaler": DummyScaler()}
+    with pytest.raises(predict.PredictException):
+        predict.MachineLearningModelHandlerScore.predict([[1]])
+
+
+def test_get_model_caches(monkeypatch):
+    predict.MachineLearningModelHandlerScore.model = None
+    monkeypatch.setattr(
+        predict.MachineLearningModelHandlerScore,
+        "load",
+        staticmethod(lambda loader: {"model": DummyModel(), "scaler": DummyScaler()}),
+    )
+    model = predict.MachineLearningModelHandlerScore.get_model(lambda path: None)
+    assert model["model"].__class__ is DummyModel
+    model2 = predict.MachineLearningModelHandlerScore.get_model(None)
+    assert model2 is model
+
+
+def test_load_model_success(tmp_path, monkeypatch):
+    dummy = tmp_path / "model.joblib"
+    dummy.write_text("data")
+    monkeypatch.setattr(predict, "MODEL_PATH", str(tmp_path))
+    monkeypatch.setattr(predict, "MODEL_NAME", "model.joblib")
+
+    def fake_loader(path):
+        assert os.path.exists(path)
+        return {"model": DummyModel(), "scaler": DummyScaler()}
+
+    model = predict.MachineLearningModelHandlerScore.load(fake_loader)
+    assert model["model"].__class__ is DummyModel
+
+
+def test_load_model_missing(tmp_path, monkeypatch):
+    monkeypatch.setattr(predict, "MODEL_PATH", str(tmp_path))
+    monkeypatch.setattr(predict, "MODEL_NAME", "missing.joblib")
+    with pytest.raises(FileNotFoundError):
+        predict.MachineLearningModelHandlerScore.load(lambda p: None)
+
+
+def test_load_model_empty(tmp_path, monkeypatch):
+    dummy = tmp_path / "model.joblib"
+    dummy.write_text("data")
+    monkeypatch.setattr(predict, "MODEL_PATH", str(tmp_path))
+    monkeypatch.setattr(predict, "MODEL_NAME", "model.joblib")
+
+    def fake_loader(path):
+        return None
+
+    with pytest.raises(predict.ModelLoadException):
+        predict.MachineLearningModelHandlerScore.load(fake_loader)