Integration Tests for NexusAI

This directory contains comprehensive integration tests for all NexusAI services.

Overview

Integration tests verify that services work together correctly in a real-world scenario. Unlike unit tests that test individual components in isolation, integration tests ensure:

• Services can communicate with each other
• Database operations work correctly
• External API integrations function properly
• Error handling works across service boundaries
• Performance meets requirements

Test Files

1. test_image_generation_integration.py

Tests for the Image Generation service:

• ✅ Full image generation workflow
• ✅ Different models (imagen-3.0-generate-001, imagen-3.0-fast-generate-001)
• ✅ Different aspect ratios (1:1, 16:9, 9:16, 4:3, 3:4)
• ✅ Safety filter levels
• ✅ Error handling (missing/empty prompts)
• ✅ Concurrent requests
• ✅ Database integration
• ✅ Performance testing

2. test_chat_integration.py

Tests for the Chat service:

• ✅ Simple chat interactions
• ✅ Conversation history management
• ✅ Tool integration (Google Search, Code Execution)
• ✅ Multiple tools simultaneously
• ✅ Different Gemini models
• ✅ Response cleaning (triple quotes removal)
• ✅ Database integration
• ✅ Error handling (empty messages, invalid history)
• ✅ Concurrent chat sessions
• ✅ Long conversation histories
• ✅ Performance testing

3. test_director_integration.py

Tests for the Director service (Video Creation):

• ✅ New movie job creation
• ✅ Movie creation with existing scenes (retake/edit flow)
• ✅ Script generation workflow
• ✅ Script approval process
• ✅ Movie status tracking
• ✅ User jobs retrieval
• ✅ External job saving (from TextToVideo)
• ✅ Database integration
• ✅ Authorization checks
• ✅ Concurrent movie creation
• ✅ Performance testing

4. test_video_generation_integration.py

Tests for the Video Generation service:

• ✅ Text-to-video workflow
• ✅ Image-to-video workflow
• ✅ Video extension workflow
• ✅ Status polling
• ✅ Save local workflow
• ✅ Different models (veo-3.1, veo-3.1-fast-generate-preview)
• ✅ Different durations (4s, 8s)
• ✅ Different resolutions (720p, 1080p)
• ✅ Different aspect ratios (16:9, 9:16, 1:1)
• ✅ Error handling (missing prompt, invalid duration, invalid operation)
• ✅ Concurrent requests
• ✅ Performance testing

5. test_document_summarization_integration.py

Tests for the Document Summarization service:

• ✅ Text summarization workflow
• ✅ PDF document summarization
• ✅ DOCX document summarization
• ✅ TXT file summarization
• ✅ Different Gemini models
• ✅ Very long documents
• ✅ Short documents
• ✅ Error handling (empty text, unsupported files, corrupted files)
• ✅ Database integration
• ✅ Concurrent requests
• ✅ Performance testing
• ✅ Summary quality checks

6. test_youtube_transcript_integration.py

Tests for the YouTube Transcript service:

• ✅ Transcript extraction workflow
• ✅ Transcript summarization
• ✅ Different YouTube URL formats
• ✅ Language-specific transcripts
• ✅ Transcripts with timestamps
• ✅ Error handling (invalid URL, video not found, no transcript, language unavailable)
• ✅ Database integration
• ✅ Long video transcripts
• ✅ Concurrent requests
• ✅ Performance testing
• ✅ Transcript formatting

7. test_integration_deployment.py

Tests for deployed environment:

• ✅ Gateway health checks
• ✅ Backend services health (via Nginx routing)
• ✅ Real API end-to-end tests (optional, requires API keys)
• ✅ Request validation and security
• ✅ CORS headers

Running the Tests

Run All Integration Tests

pytest tests/integration -v

Run Specific Test File

pytest tests/integration/test_image_generation_integration.py -v
pytest tests/integration/test_chat_integration.py -v
pytest tests/integration/test_director_integration.py -v
pytest tests/integration/test_video_generation_integration.py -v
pytest tests/integration/test_document_summarization_integration.py -v
pytest tests/integration/test_youtube_transcript_integration.py -v
pytest tests/integration/test_integration_deployment.py -v

Run Specific Test Class

pytest tests/integration/test_image_generation_integration.py::TestImageGenerationIntegration -v

Run Specific Test Method

pytest tests/integration/test_chat_integration.py::TestChatIntegration::test_simple_chat_workflow -v

Run with Coverage

pytest tests/integration --cov=. --cov-report=html

Run with Detailed Output

pytest tests/integration -vv --tb=long

Prerequisites

For Local Testing (Mocked APIs)

• Python 3.8+
• All dependencies installed: pip install . (uses pyproject.toml)
• No API keys required (tests use mocks)

For Real API Testing

• Set environment variables:

  export GEMINI_API_KEY=your_key_here
  export GOOGLE_CLOUD_PROJECT=your_project_id

• Services running (for deployment tests):

  docker-compose up
  # OR
  ./start_all.ps1

Test Structure

Each integration test follows this pattern:

class TestServiceIntegration:
    def test_health_endpoint(self):
        """Verify service is running"""
        
    def test_main_workflow(self, mock_api):
        """Test the primary use case"""
        
    def test_edge_cases(self):
        """Test boundary conditions"""
        
    def test_error_handling(self):
        """Test failure scenarios"""
        
    def test_database_integration(self):
        """Verify database operations"""
        
    def test_performance(self):
        """Ensure acceptable response times"""

Mocking Strategy

Integration tests use mocks for external APIs to:

• ✅ Avoid API costs during testing
• ✅ Ensure consistent test results
• ✅ Test error scenarios that are hard to reproduce
• ✅ Speed up test execution

Mocked components:

• Gemini API (genai)
• Imagen API (imagen)
• Veo API (client)
• YouTube Transcript API (YouTubeTranscriptApi)
• Database (db)
• Storage (storage)

CI/CD Integration

To integrate these tests into your CI/CD pipeline:

Cloud Build (cloudbuild.yaml)

steps:
  - name: 'python:3.11'
    entrypoint: 'bash'
    args:
      - '-c'
      - |
        pip install .
        pytest tests/integration -v --tb=short

GitHub Actions

- name: Run Integration Tests
  run: |
    pip install .
    pytest tests/integration -v

Best Practices

1. Isolation: Each test should be independent and not rely on other tests
2. Cleanup: Use fixtures to clean up test data after tests
3. Mocking: Mock external APIs to avoid costs and ensure reliability
4. Assertions: Use clear, specific assertions with helpful error messages
5. Documentation: Document what each test is verifying
6. Performance: Set reasonable timeout expectations
7. Error Cases: Test both success and failure scenarios

Troubleshooting

Tests Fail with "Connection Refused"

• Ensure services are running: docker-compose up or ./start_all.ps1
• Check port availability (8080, 8001-8006)

Tests Fail with "Module Not Found"

• Install dependencies: pip install . (or uv sync if using uv)
• Ensure you're in the project root directory

Tests Fail with "API Key Error"

• For mocked tests: This shouldn't happen (check fixtures)
• For real API tests: Set GEMINI_API_KEY environment variable

Slow Test Execution

• Use mocked tests for development
• Run specific test files instead of all tests
• Use -k flag to run specific tests: pytest -k "test_simple"

Contributing

When adding new features:

1. Add corresponding integration tests
2. Follow the existing test structure
3. Use appropriate mocks for external APIs
4. Document what the test verifies
5. Ensure tests pass before submitting PR

Test Coverage Goals

• Unit Tests: 80%+ code coverage
• Integration Tests: Cover all major workflows
• E2E Tests: Cover critical user journeys

Next Steps

1. ✅ Run all integration tests: pytest tests/integration -v
2. ✅ Check coverage: pytest tests/integration --cov=. --cov-report=html
3. ✅ Add to CI/CD pipeline
4. ✅ Monitor test execution times
5. ✅ Add more edge case tests as needed

Contact

For questions or issues with integration tests, please open an issue or contact the development team.