Integration Tests for Microsoft Fabric Casino/Gaming POC¶

This directory contains comprehensive integration tests for validating the data generation, pipeline transformations, schema compliance, and regulatory requirements of the Casino/Gaming proof-of-concept.

Test Structure¶

integration_tests/
├── conftest.py                    # Shared fixtures and configuration
├── test_generator_integration.py  # Generator integration tests
├── test_data_pipeline.py          # Data pipeline transformation tests
├── test_schema_compliance.py      # JSON schema validation tests
├── test_compliance_rules.py       # Business/regulatory compliance tests
├── test_streaming_integration.py  # Streaming functionality tests
├── test_notebook_imports.py       # Notebook dependency tests
└── README.md                      # This file

Test Categories¶

1. Generator Integration Tests (test_generator_integration.py)¶

Tests that verify all data generators work together correctly: - All generators can be instantiated together - Player IDs are compatible across generators for cross-referencing - Date ranges are consistent across generators - Seed reproducibility for deterministic test runs

2. Data Pipeline Tests (test_data_pipeline.py)¶

Tests for medallion architecture transformations: - Bronze -> Silver: Null filtering, type casting, standardization, deduplication, data quality scoring - Silver -> Gold: Aggregations, KPI calculations, net win, hold percentage, player metrics - Data type consistency through layers - Null handling behavior

3. Schema Compliance Tests (test_schema_compliance.py)¶

Tests validating generated data against JSON schemas: - Required field presence - Field type validation (string, number, datetime) - Enum value validation (event types, zones, loyalty tiers) - Pattern matching (machine_id, player_id formats) - Constraint validation (minimum values, ranges)

4. Compliance Rules Tests (test_compliance_rules.py)¶

Tests for regulatory compliance requirements: - CTR Threshold: $10,000 triggers Currency Transaction Report - W-2G Threshold: $1,200 for slots, $600 for table games (300:1 odds) - SAR Detection: Suspicious Activity Report pattern detection - PII Masking: SSN hashing, name masking, no raw PII exposure

5. Streaming Integration Tests (test_streaming_integration.py)¶

Tests for real-time streaming functionality: - EventHubProducer initialization and configuration - Event generation and JSON serialization - Rate limiting and max events constraints - Graceful shutdown handling - Callback functionality

6. Notebook Import Tests (test_notebook_imports.py)¶

Tests for notebook dependencies: - Import statement parsing and validation - Standard library availability - No circular dependencies between notebooks - Layer ordering (Bronze -> Silver -> Gold) - Required packages documentation

Running Tests¶

Run All Integration Tests¶

cd E:\Repos\GitHub\MyDemoRepos\Suppercharge_Microsoft_Fabric
pytest validation/integration_tests/ -v

Run by Marker¶

# Integration tests only
pytest validation/integration_tests/ -m integration -v

# Compliance tests only
pytest validation/integration_tests/ -m compliance -v

# Schema tests only
pytest validation/integration_tests/ -m schema -v

# Streaming tests only
pytest validation/integration_tests/ -m streaming -v

# Pipeline tests only
pytest validation/integration_tests/ -m pipeline -v

Run Specific Test File¶

# Generator tests
pytest validation/integration_tests/test_generator_integration.py -v

# Compliance rules tests
pytest validation/integration_tests/test_compliance_rules.py -v

Skip Slow Tests¶

pytest validation/integration_tests/ -m "not slow" -v

Run with Coverage¶

pytest validation/integration_tests/ --cov=data_generation --cov-report=html

Test Markers¶

The tests use the following pytest markers:

Fixtures¶

Session-Scoped Fixtures¶

These fixtures are created once per test session for efficiency:

Schema Fixtures¶

Validation Fixtures¶

Compliance Thresholds¶

The tests validate the following regulatory thresholds:

CTR (Currency Transaction Report)¶

• Threshold: $10,000 aggregate cash transactions per gaming day
• Required fields: Amount, player ID, transaction date

W-2G (Gambling Winnings)¶

• Slots/Video Poker: $1,200
• Keno: $1,500
• Bingo: $1,200
• Poker Tournament: $5,000
• Table Games: $600 (at 300:1 odds or greater)

SAR (Suspicious Activity Report)¶

• Structuring: Multiple transactions just below $10,000
• Required: Suspicion category and narrative

PII Protection¶

• SSN: SHA-256 hashed (64 hex characters)
• SSN Display: Masked format XXX-XX-####
• Names: Masked when PII disabled
• Addresses: Masked when PII disabled

Adding New Tests¶

1. Add Test Class¶

class TestNewFeature:
    """Tests for new feature."""

    def test_feature_behavior(self, sample_slot_data):
        """Test specific behavior."""
        # Arrange
        data = sample_slot_data.copy()

        # Act
        result = some_transformation(data)

        # Assert
        assert expected_condition

2. Use Appropriate Markers¶

@pytest.mark.compliance
@pytest.mark.slow
def test_complex_compliance_rule(self, sample_compliance_data):
    ...

3. Add Fixtures if Needed¶

Add to conftest.py:

@pytest.fixture(scope="session")
def new_fixture():
    """Description of fixture."""
    return setup_value

Dependencies¶

Required packages for running tests: - pytest >= 7.0 - pytest-asyncio (for async tests) - pandas - numpy - faker

Optional: - pytest-cov (for coverage reports) - azure-eventhub (for full streaming tests)

Troubleshooting¶

Import Errors¶

Ensure the data_generation directory is in PYTHONPATH:

set PYTHONPATH=%PYTHONPATH%;E:\Repos\GitHub\MyDemoRepos\Suppercharge_Microsoft_Fabric\data_generation

Fixture Not Found¶

Verify conftest.py is in the test directory and contains the required fixture.

Schema Not Found¶

Ensure JSON schemas exist in data_generation/schemas/.

Slow Test Timeout¶

Use -s flag to see output and increase timeout:

pytest --timeout=300 -s