add async to rate limiting testing

routers/predict.py

@@ -91,7 +91,7 @@ def prediction(dataset, sent_model, text_len):
     return predictions.item()
 
 @router.post("/", status_code=status.HTTP_200_OK, response_model=PredictResponse)
-@limiter.limit("2/hour")  # Limit to 2 requests per minute
+@limiter.limit("2/minute")  # Limit to 2 requests per minute
 async def predict_sentiment(request:Request, data: TextStr):
     """
     Predict the sentiment of the input text using a pre-loaded machine learning model.

tests/test_routers/test_predict.py

@@ -25,6 +25,7 @@
 
 import os
 
+import httpx
 import tensorflow as tf
 import tensorflow_hub as hub
 import tensorflow_text as text  # pylint: disable=unused-import
@@ -37,7 +38,7 @@
 # Create a test FastAPI app
 app = FastAPI()
 app.include_router(predict.router)
-
+    
 @pytest.fixture
 def client():
     """
@@ -71,6 +72,35 @@ def client():
     with TestClient(app) as c:
         yield c
 
+import pytest_asyncio
+import httpx
+
+@pytest_asyncio.fixture
+async def async_client():
+    """
+    Asynchronous fixture to set up a test client for the FastAPI application with 
+    a loaded sentiment analysis model.
+
+    This fixture initializes the FastAPI application by loading the sentiment analysis 
+    model from the specified directory and sets it in the app's state. It provides an 
+    async HTTP client for testing FastAPI endpoints.
+    """
+    # Load the model as before
+    model_path = os.path.join(os.path.dirname(__file__), '../../nlp-models/saved-model/sent-model')
+
+    ml_models = tf.keras.models.load_model(
+        model_path,
+        custom_objects={'KerasLayer': hub.KerasLayer}
+    )
+    ml_models.trainable = False
+
+    # Assign the model to the FastAPI app state
+    app.state.sentiment_analysis_model = ml_models
+
+    # Create an asynchronous client for testing
+    async with httpx.AsyncClient(app=app, base_url="http://127.0.0.1:8000/") as client:
+        yield client
+
 def test_model_loaded(client):
     """
     Tests the testclient for loaded model.
@@ -145,7 +175,8 @@ def test_predict_sentiment_model_not_loaded(client):
     assert response.status_code == 500
     assert response.json() == {"detail": "Model not loaded properly"}
 
-def test_rate_limiting(client):
+@pytest.mark.asyncio
+async def test_rate_limiting(async_client):
     """
     Test that the /predict route is limited to 2 requests per minute.
     The 3rd request should return a 429 Too Many Requests error.
@@ -161,14 +192,13 @@ def test_rate_limiting(client):
     """
     test_data = {"texts": "I recently stayed at a hotel that was highly disappointing. The room was dirty, and the staff were unhelpful and rude."}
 
-
     # Make 2 successful requests within the rate limit
     for _ in range(2):
-        response = client.post("/", json=test_data)
+        response = await async_client.post("/", json=test_data)
         assert response.status_code == 200, "Expected status code 200 for valid requests"
 
     # The 3rd request should hit the rate limit and return status 429
-    response = client.post("/", json=test_data)
+    response = await async_client.post("/", json=test_data)
     assert response.status_code == 429, "Expected status code 429 after rate limit exceeded"
-    assert response.json() == {'detail': '2 per 1 hour'}, \
+    assert response.json() == {'detail': '2 per 1 minute'}, \
         "Expected rate limit exceeded error message"