tests/test_api.py

"""Tests for the Spik API endpoints."""
import os
import pytest
from fastapi.testclient import TestClient
from app.main import app

# Test client
client = TestClient(app)

# Test data
TEST_TEXT = "Hello, this is a test message."
TEST_VOICE = "Deepgram Aura2"

@pytest.fixture(scope="module")
def test_audio_file():
    """Generate a test audio file."""
    # This would be replaced with an actual test audio file in a real test
    test_file = "test_audio.wav"
    with open(test_file, "wb") as f:
        # Write a minimal WAV header (44 bytes)
        f.write(b'RIFF$\xbc\x00\x00\x00WAVEfmt \x10\x00\x00\x00\x01\x00\x01\x00')
        f.write(b'\x44\xac\x00\x00\x88X\x01\x00\x02\x00\x10\x00data\x00\x00\x00\x00')
    yield test_file
    # Cleanup
    if os.path.exists(test_file):
        os.remove(test_file)

def test_health_check():
    """Test the health check endpoint."""
    response = client.get("/api/health")
    assert response.status_code == 200
    assert response.json()["status"] == "healthy"

@pytest.mark.skipif(
    not os.getenv("DEEPGRAM_API_KEY"),
    reason="DEEPGRAM_API_KEY not set in environment"
)
def test_tts_endpoint():
    """Test the text-to-speech endpoint."""
    response = client.post(
        "/api/tts",
        json={"text": TEST_TEXT, "voice": TEST_VOICE}
    )
    assert response.status_code == 200
    data = response.json()
    assert "audio_url" in data
    assert data["audio_url"].startswith("/output/tts_")
    assert "metadata" in data
    assert data["metadata"]["voice"] == TEST_VOICE

@pytest.mark.skipif(
    not os.getenv("DEEPGRAM_API_KEY"),
    reason="DEEPGRAM_API_KEY not set in environment"
)
def test_stt_endpoint(test_audio_file):
    """Test the speech-to-text endpoint with file upload."""
    with open(test_audio_file, "rb") as f:
        response = client.post(
            "/api/stt",
            files={"audio": ("test.wav", f, "audio/wav")},
            data={"language": "en", "model": "nova-3"}
        )
    assert response.status_code == 200
    data = response.json()
    assert "transcript" in data
    # We can't assert the exact transcript since it depends on the audio content
    assert isinstance(data["transcript"], str)

@pytest.mark.skipif(
    not os.getenv("DEEPGRAM_API_KEY"),
    reason="DEEPGRAM_API_KEY not set in environment"
)
def test_stt_with_url():
    """Test the speech-to-text endpoint with a URL."""
    # Using a short test audio URL - in a real test, this would be a URL to a test audio file
    test_audio_url = "https://dpgr.am/spacewalk.wav"
    response = client.post(
        "/api/stt",
        data={
            "audio_url": test_audio_url,
            "language": "en",
            "model": "nova-3"
        }
    )
    assert response.status_code == 200
    data = response.json()
    assert "transcript" in data
    assert isinstance(data["transcript"], str)

def test_tts_validation():
    """Test validation of the TTS endpoint."""
    # Test empty text
    response = client.post("/api/tts", json={"text": "", "voice": TEST_VOICE})
    assert response.status_code == 400
    
    # Test invalid voice
    response = client.post(
        "/api/tts",
        json={"text": TEST_TEXT, "voice": "Invalid Voice"}
    )
    assert response.status_code == 400

def test_stt_validation():
    """Test validation of the STT endpoint."""
    # Test no audio or URL provided
    response = client.post("/api/stt")
    assert response.status_code == 400
    
    # Test invalid content type
    response = client.post(
        "/api/stt",
        files={"audio": ("test.txt", b"not an audio file", "text/plain")}
    )
    assert response.status_code == 400