foxhunt/adaptive-strategy/tests/algorithm_comprehensive.rs

//! Comprehensive algorithm tests for adaptive-strategy crate
//!
//! This test suite provides extensive coverage for:
//! - Strategy signal generation algorithms
//! - Position sizing logic (Kelly, PPO, risk parity)
//! - Ensemble model coordination and voting
//! - Performance tracking and metrics
//! - Model factory and registry operations
//! - Regime detection and adaptation
//!
//! Target: Increase coverage from 40-50% → 75%+

use adaptive_strategy::config::{
    AdaptiveStrategyConfig, ExecutionAlgorithm, ModelConfig, PositionSizingMethod,
    RegimeDetectionMethod,
};
use adaptive_strategy::ensemble::EnsembleCoordinator;
use adaptive_strategy::models::{
    ModelFactory, ModelMetadata, ModelRegistry, ModelTrait, TrainingData,
};
use adaptive_strategy::risk::{
    KellyConfig, KellyPositionSizer, PortfolioRiskMetrics, PositionRiskMetrics,
    PositionSizeRecommendation, RiskManager,
};
use adaptive_strategy::{AdaptiveStrategy, PerformanceMetrics, StrategyState};
use common::{MarketRegime, Position};
use rust_decimal::Decimal;
use std::collections::HashMap;

// ============================================================================
// STRATEGY ALGORITHM TESTS (10 tests)
// ============================================================================

#[tokio::test]
async fn test_adaptive_strategy_creation_with_default_config() {
    let config = AdaptiveStrategyConfig::default();
    let strategy = AdaptiveStrategy::new(config).await;
    assert!(strategy.is_ok(), "Strategy creation should succeed");

    let strategy = strategy.unwrap();
    let state = strategy.get_state().await;
    assert!(!state.active, "Initial strategy should be inactive");
    assert_eq!(state.current_regime, "unknown");
}

#[tokio::test]
async fn test_adaptive_strategy_state_transitions() {
    let config = AdaptiveStrategyConfig::default();
    let strategy = AdaptiveStrategy::new(config).await.unwrap();

    // Initial state
    let initial_state = strategy.get_state().await;
    assert!(!initial_state.active);

    // Strategy lifecycle would be tested here if start() didn't run indefinitely
    // In production, we'd use a mock or time-limited execution
}

#[tokio::test]
async fn test_strategy_performance_metrics_initialization() {
    let metrics = PerformanceMetrics::default();

    assert_eq!(metrics.sharpe_ratio, 0.0);
    assert_eq!(metrics.max_drawdown, 0.0);
    assert_eq!(metrics.total_return, 0.0);
    assert_eq!(metrics.win_rate, 0.0);
    assert_eq!(metrics.trade_count, 0);
}

#[tokio::test]
async fn test_strategy_config_update() {
    let config = AdaptiveStrategyConfig::default();
    let mut strategy = AdaptiveStrategy::new(config.clone()).await.unwrap();

    // Create modified config
    let mut new_config = config;
    new_config.risk.max_leverage = 3.0;

    let result = strategy.update_config(new_config).await;
    assert!(result.is_ok(), "Config update should succeed");
}

#[tokio::test]
async fn test_strategy_state_serialization() {
    let state = StrategyState {
        active: true,
        current_regime: "bull".to_string(),
        model_weights: HashMap::from([
            ("mamba2".to_string(), 0.4),
            ("tlob".to_string(), 0.6),
        ]),
        last_update: chrono::Utc::now(),
        performance: PerformanceMetrics::default(),
    };

    let serialized = serde_json::to_string(&state);
    assert!(serialized.is_ok(), "State should be serializable");

    let deserialized: Result<StrategyState, _> = serde_json::from_str(&serialized.unwrap());
    assert!(deserialized.is_ok(), "State should be deserializable");
}

#[tokio::test]
async fn test_strategy_with_kelly_position_sizing() {
    let mut config = AdaptiveStrategyConfig::default();
    config.risk.position_sizing_method = PositionSizingMethod::Kelly;
    config.risk.kelly_fraction = 0.25;

    let strategy = AdaptiveStrategy::new(config).await;
    assert!(strategy.is_ok(), "Strategy with Kelly sizing should create successfully");
}

#[tokio::test]
async fn test_strategy_with_ppo_position_sizing() {
    let mut config = AdaptiveStrategyConfig::default();
    config.risk.position_sizing_method = PositionSizingMethod::PPO;

    let strategy = AdaptiveStrategy::new(config).await;
    assert!(strategy.is_ok(), "Strategy with PPO sizing should create successfully");
}

#[tokio::test]
async fn test_strategy_with_custom_execution_algorithm() {
    let mut config = AdaptiveStrategyConfig::default();
    config.execution.algorithm = ExecutionAlgorithm::VWAP;
    config.execution.max_slippage_bps = 5.0;

    let strategy = AdaptiveStrategy::new(config).await;
    assert!(strategy.is_ok(), "Strategy with VWAP should create successfully");
}

#[tokio::test]
async fn test_strategy_with_hmm_regime_detection() {
    let mut config = AdaptiveStrategyConfig::default();
    config.regime.detection_method = RegimeDetectionMethod::HMM;
    config.regime.lookback_window = 252;

    let strategy = AdaptiveStrategy::new(config).await;
    assert!(strategy.is_ok(), "Strategy with HMM regime detection should succeed");
}

#[tokio::test]
async fn test_strategy_with_multiple_models() {
    let mut config = AdaptiveStrategyConfig::default();
    config.ensemble.models = vec![
        ModelConfig {
            id: "mamba2".to_string(),
            name: "mamba2_model".to_string(),
            model_type: "mamba2".to_string(),
            parameters: serde_json::Value::Null,
            initial_weight: 0.25,
            enabled: true,
        },
        ModelConfig {
            id: "tlob".to_string(),
            name: "tlob_model".to_string(),
            model_type: "tlob".to_string(),
            parameters: serde_json::Value::Null,
            initial_weight: 0.25,
            enabled: true,
        },
        ModelConfig {
            id: "lstm".to_string(),
            name: "lstm_model".to_string(),
            model_type: "lstm".to_string(),
            parameters: serde_json::Value::Null,
            initial_weight: 0.25,
            enabled: true,
        },
        ModelConfig {
            id: "transformer".to_string(),
            name: "transformer_model".to_string(),
            model_type: "transformer".to_string(),
            parameters: serde_json::Value::Null,
            initial_weight: 0.25,
            enabled: true,
        },
    ];

    let strategy = AdaptiveStrategy::new(config).await;
    assert!(strategy.is_ok(), "Strategy with 4 models should create successfully");
}

// ============================================================================
// POSITION SIZING ALGORITHM TESTS (10 tests)
// ============================================================================

#[tokio::test]
async fn test_kelly_position_sizer_creation() {
    let config = KellyConfig {
        max_fraction: 0.25,
        min_fraction: 0.01,
        lookback_period: 252,
        confidence_threshold: 0.6,
        volatility_adjustment: true,
        drawdown_protection: true,
        dynamic_risk_scaling: true,
        max_concentration: 0.20,
        correlation_adjustment: 0.85,
        base_kelly: 0.25,
    };

    let sizer = KellyPositionSizer::new(config);
    assert!(sizer.is_ok(), "Kelly sizer should create successfully");
}

#[tokio::test]
async fn test_kelly_position_sizing_calculation() {
    let config = KellyConfig::default();
    let mut sizer = KellyPositionSizer::new(config).unwrap();

    let historical_returns = vec![
        0.05, -0.02, 0.08, -0.03, 0.06, -0.01, 0.04, -0.02,
        0.07, -0.01, 0.03, -0.04, 0.09, -0.02, 0.05, -0.03,
    ];

    let mut market_data = adaptive_strategy::risk::MarketData {
        prices: HashMap::new(),
        volatilities: HashMap::new(),
        correlations: HashMap::new(),
        timestamp: chrono::Utc::now(),
        volatility_index: Some(20.0),
        sentiment_indicators: HashMap::new(),
    };

    market_data.prices.insert("AAPL".to_string(), 150.0);
    market_data.volatilities.insert("AAPL".to_string(), 0.25);

    let recommendation = sizer
        .calculate_position_size("AAPL", 0.08, 0.75, &historical_returns, &market_data)
        .await;

    assert!(recommendation.is_ok(), "Kelly calculation should succeed");
    let rec = recommendation.unwrap();
    assert!(rec.recommended_fraction >= 0.0 && rec.recommended_fraction <= 0.25);
    assert!(rec.confidence > 0.0 && rec.confidence <= 1.0);
}

#[tokio::test]
async fn test_kelly_with_high_volatility_regime() {
    let config = KellyConfig::default();
    let mut sizer = KellyPositionSizer::new(config).unwrap();

    // Update to high volatility regime
    sizer
        .update_market_regime(adaptive_strategy::regime::MarketRegime::HighVolatility)
        .await
        .unwrap();

    let historical_returns = vec![0.05; 20];
    let mut market_data = adaptive_strategy::risk::MarketData {
        prices: HashMap::new(),
        volatilities: HashMap::new(),
        correlations: HashMap::new(),
        timestamp: chrono::Utc::now(),
        volatility_index: Some(40.0), // High VIX
        sentiment_indicators: HashMap::new(),
    };

    market_data.prices.insert("SPY".to_string(), 450.0);
    market_data.volatilities.insert("SPY".to_string(), 0.35); // High volatility

    let recommendation = sizer
        .calculate_position_size("SPY", 0.05, 0.7, &historical_returns, &market_data)
        .await
        .unwrap();

    // In high volatility, Kelly should recommend smaller positions
    assert!(
        recommendation.recommended_fraction < 0.15,
        "High volatility should reduce position size"
    );
}

#[tokio::test]
async fn test_kelly_with_drawdown_protection() {
    let config = KellyConfig {
        drawdown_protection: true,
        max_fraction: 0.25,
        ..Default::default()
    };

    let sizer = KellyPositionSizer::new(config);
    assert!(sizer.is_ok());

    // Drawdown protection should reduce position sizes during losses
    // This would be tested with actual drawdown tracking in production
}

#[tokio::test]
async fn test_fixed_fractional_position_sizing() {
    let mut config = AdaptiveStrategyConfig::default();
    config.risk.position_sizing_method = PositionSizingMethod::FixedFractional(0.1);

    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let recommendation = risk_manager
        .calculate_position_size("MSFT", 0.06, 0.8, 300.0)
        .await;

    assert!(recommendation.is_ok());
    let rec = recommendation.unwrap();
    assert!(rec.size > 0.0, "Should recommend non-zero position");
}

#[tokio::test]
async fn test_risk_parity_position_sizing() {
    let mut config = AdaptiveStrategyConfig::default();
    config.risk.position_sizing_method = PositionSizingMethod::RiskParity;

    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let recommendation = risk_manager
        .calculate_position_size("GOOGL", 0.07, 0.85, 140.0)
        .await;

    assert!(recommendation.is_ok());
}

#[tokio::test]
async fn test_volatility_target_position_sizing() {
    let mut config = AdaptiveStrategyConfig::default();
    config.risk.position_sizing_method = PositionSizingMethod::VolatilityTarget;

    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let recommendation = risk_manager
        .calculate_position_size("TSLA", 0.10, 0.65, 250.0)
        .await;

    assert!(recommendation.is_ok());
}

#[tokio::test]
async fn test_position_size_with_risk_limits() {
    let mut config = AdaptiveStrategyConfig::default();
    config.risk.max_position_size = 0.05; // 5% limit
    config.risk.max_leverage = 1.5;

    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let recommendation = risk_manager
        .calculate_position_size("NVDA", 0.15, 0.9, 500.0)
        .await
        .unwrap();

    // Position should be constrained by risk limits
    assert!(
        recommendation.size <= recommendation.max_allowed_size,
        "Position should not exceed max allowed"
    );
}

#[tokio::test]
async fn test_position_sizing_risk_metrics() {
    let config = AdaptiveStrategyConfig::default();
    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let recommendation = risk_manager
        .calculate_position_size("AMZN", 0.08, 0.75, 180.0)
        .await
        .unwrap();

    // Verify risk metrics are calculated
    assert!(recommendation.risk_metrics.expected_return > 0.0);
    assert!(recommendation.risk_metrics.expected_volatility >= 0.0);
    assert!(recommendation.risk_metrics.var_95 >= 0.0);
    assert!(recommendation.risk_metrics.cvar_95 >= recommendation.risk_metrics.var_95);
}

#[tokio::test]
async fn test_position_size_recommendation_serialization() {
    let recommendation = PositionSizeRecommendation {
        size: 100.0,
        confidence: 0.85,
        max_allowed_size: 150.0,
        method: "Kelly".to_string(),
        risk_metrics: PositionRiskMetrics {
            expected_return: 0.08,
            expected_volatility: 0.20,
            sharpe_ratio: 0.4,
            var_95: 20.0,
            cvar_95: 25.0,
            max_loss: 100.0,
        },
        timestamp: chrono::Utc::now(),
    };

    let serialized = serde_json::to_string(&recommendation);
    assert!(serialized.is_ok());

    let deserialized: Result<PositionSizeRecommendation, _> =
        serde_json::from_str(&serialized.unwrap());
    assert!(deserialized.is_ok());
}

// ============================================================================
// ENSEMBLE MODEL COORDINATION TESTS (5 tests)
// ============================================================================

#[tokio::test]
async fn test_ensemble_coordinator_creation() {
    let config = AdaptiveStrategyConfig::default();
    let coordinator = EnsembleCoordinator::new(&config).await;
    assert!(coordinator.is_ok(), "Ensemble coordinator should create successfully");
}

#[tokio::test]
async fn test_ensemble_prediction_generation() {
    let config = AdaptiveStrategyConfig::default();
    let coordinator = EnsembleCoordinator::new(&config).await.unwrap();

    let features = vec![1.0, 2.0, 3.0, 4.0, 5.0];
    let horizon = chrono::Duration::seconds(60);

    let prediction = coordinator.predict(&features, horizon).await;
    assert!(prediction.is_ok(), "Ensemble prediction should succeed");

    let pred = prediction.unwrap();
    assert!(
        pred.confidence >= 0.0 && pred.confidence <= 1.0,
        "Confidence should be in [0,1]"
    );
    assert!(!pred.model_contributions.is_empty(), "Should have model contributions");
}

#[tokio::test]
async fn test_ensemble_weight_updates() {
    let config = AdaptiveStrategyConfig::default();
    let coordinator = EnsembleCoordinator::new(&config).await.unwrap();

    let initial_weights = coordinator.get_weights().await;
    assert!(!initial_weights.is_empty(), "Should have initial weights");

    // Update weights
    let result = coordinator.update_weights_default().await;
    assert!(result.is_ok(), "Weight update should succeed");

    let updated_weights = coordinator.get_weights().await;
    assert_eq!(initial_weights.len(), updated_weights.len());
}

#[tokio::test]
async fn test_ensemble_outcome_recording() {
    let config = AdaptiveStrategyConfig::default();
    let coordinator = EnsembleCoordinator::new(&config).await.unwrap();

    let timestamp = chrono::Utc::now();
    let actual_outcome = 0.05;

    let result = coordinator.record_outcome_legacy(timestamp, actual_outcome).await;
    assert!(result.is_ok(), "Outcome recording should succeed");
}

#[tokio::test]
async fn test_ensemble_performance_tracking() {
    let config = AdaptiveStrategyConfig::default();
    let coordinator = EnsembleCoordinator::new(&config).await.unwrap();

    let performance = coordinator.get_performance().await;
    assert_eq!(performance.accuracy().len(), 0, "Initial accuracy should be empty");
}

// ============================================================================
// MODEL FACTORY AND REGISTRY TESTS (5 tests)
// ============================================================================

#[tokio::test]
async fn test_model_factory_available_models() {
    let models = ModelFactory::available_models();
    assert!(models.contains(&"lstm"), "Should support LSTM");
    assert!(models.contains(&"transformer"), "Should support Transformer");
    assert!(models.contains(&"random_forest"), "Should support Random Forest");
    assert!(models.contains(&"xgboost"), "Should support XGBoost");
    assert!(models.contains(&"tlob"), "Should support TLOB");
}

#[tokio::test]
async fn test_model_factory_creation() {
    let config = adaptive_strategy::models::ModelConfig::default();

    let model = ModelFactory::create_model("lstm", "test_lstm".to_string(), config).await;
    assert!(model.is_ok(), "LSTM model creation should succeed");

    let model = model.unwrap();
    assert_eq!(model.name(), "test_lstm");
    assert_eq!(model.model_type(), "lstm");
}

#[tokio::test]
async fn test_model_registry_operations() {
    let mut registry = ModelRegistry::new();
    assert_eq!(registry.list_models().len(), 0, "Registry should start empty");

    let config = adaptive_strategy::models::ModelConfig::default();
    let model = ModelFactory::create_model("mock", "test_model".to_string(), config)
        .await
        .unwrap();

    registry.register(model);
    assert_eq!(registry.list_models().len(), 1);
    assert!(registry.get("test_model").is_some());

    let removed = registry.remove("test_model");
    assert!(removed.is_some());
    assert_eq!(registry.list_models().len(), 0);
}

#[tokio::test]
async fn test_model_training_data_validation() {
    let features = vec![vec![1.0, 2.0, 3.0]; 10];
    let targets = vec![0.5; 10];
    let feature_names = vec!["f1".to_string(), "f2".to_string(), "f3".to_string()];

    let data = TrainingData::new(features, targets, feature_names);
    assert!(data.validate().is_ok(), "Valid data should pass validation");
    assert_eq!(data.len(), 10);
    assert!(!data.is_empty());
}

#[tokio::test]
async fn test_model_training_data_invalid() {
    let features = vec![vec![1.0, 2.0]; 5];
    let targets = vec![0.5; 3]; // Mismatched length
    let feature_names = vec!["f1".to_string(), "f2".to_string()];

    let data = TrainingData::new(features, targets, feature_names);
    assert!(data.validate().is_err(), "Invalid data should fail validation");
}

// ============================================================================
// RISK MANAGEMENT INTEGRATION TESTS (5 tests)
// ============================================================================

#[tokio::test]
async fn test_risk_manager_position_update() {
    let config = AdaptiveStrategyConfig::default();
    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let position = Position {
        id: uuid::Uuid::new_v4(),
        symbol: "AAPL".into(),
        quantity: Decimal::from(100),
        avg_price: Decimal::from(150),
        avg_cost: Decimal::from(150),
        basis: Decimal::from(15000),
        average_price: Decimal::from(150),
        market_value: Decimal::from(15000),
        unrealized_pnl: Decimal::ZERO,
        realized_pnl: Decimal::ZERO,
        created_at: chrono::Utc::now(),
        updated_at: chrono::Utc::now(),
        last_updated: chrono::Utc::now(),
        current_price: Some(Decimal::from(155)),
        notional_value: Decimal::from(15500),
        margin_requirement: Decimal::from(1550),
    };

    let result = risk_manager.update_position(position).await;
    assert!(result.is_ok(), "Position update should succeed");
}

#[tokio::test]
async fn test_risk_manager_portfolio_metrics() {
    let config = AdaptiveStrategyConfig::default();
    let risk_manager = RiskManager::new(config.risk).unwrap();

    let metrics = risk_manager.get_portfolio_risk_metrics().await;
    assert!(metrics.is_ok(), "Portfolio metrics should be calculated");

    let metrics = metrics.unwrap();
    assert!(metrics.portfolio_var >= 0.0);
    assert!(metrics.leverage >= 0.0);
    assert!(metrics.current_drawdown >= 0.0);
}

#[tokio::test]
async fn test_risk_manager_trade_risk_check() {
    let config = AdaptiveStrategyConfig::default();
    let risk_manager = RiskManager::new(config.risk).unwrap();

    let passes = risk_manager.check_trade_risk("MSFT", 50.0, 300.0).await;
    assert!(passes.is_ok(), "Trade risk check should execute");
}

#[tokio::test]
async fn test_risk_manager_limits_status() {
    let config = AdaptiveStrategyConfig::default();
    let risk_manager = RiskManager::new(config.risk).unwrap();

    let status = risk_manager.get_risk_limits_status().await;
    assert!(status.is_ok(), "Should retrieve risk limits status");

    let status = status.unwrap();
    assert!(status.contains_key("leverage") || status.contains_key("drawdown"));
}

#[tokio::test]
async fn test_risk_manager_market_regime_update() {
    let config = AdaptiveStrategyConfig::default();
    let mut risk_manager = RiskManager::new(config.risk).unwrap();

    let result = risk_manager.update_market_regime(MarketRegime::Bull).await;
    assert!(result.is_ok(), "Market regime update should succeed");

    let result = risk_manager.update_market_regime(MarketRegime::HighVolatility).await;
    assert!(result.is_ok(), "High volatility regime update should succeed");
}

// ============================================================================
// PERFORMANCE METRICS AND TRACKING TESTS (5 tests)
// ============================================================================

#[tokio::test]
async fn test_portfolio_risk_metrics_serialization() {
    let metrics = PortfolioRiskMetrics {
        portfolio_var: 0.015,
        portfolio_cvar: 0.019,
        leverage: 1.8,
        current_drawdown: 0.03,
        max_drawdown: 0.05,
        sharpe_ratio: 1.5,
        sortino_ratio: 1.8,
        beta: Some(0.95),
        concentration_risk: 0.15,
        timestamp: chrono::Utc::now(),
    };

    let serialized = serde_json::to_string(&metrics);
    assert!(serialized.is_ok());

    let deserialized: Result<PortfolioRiskMetrics, _> =
        serde_json::from_str(&serialized.unwrap());
    assert!(deserialized.is_ok());
}

#[tokio::test]
async fn test_position_risk_metrics_calculation() {
    let metrics = PositionRiskMetrics {
        expected_return: 0.08,
        expected_volatility: 0.20,
        sharpe_ratio: 0.4,
        var_95: 1000.0,
        cvar_95: 1280.0,
        max_loss: 5000.0,
    };

    assert!(metrics.sharpe_ratio > 0.0);
    assert!(metrics.cvar_95 > metrics.var_95, "CVaR should exceed VaR");
    assert!(metrics.max_loss >= metrics.cvar_95, "Max loss should be highest");
}

#[tokio::test]
async fn test_performance_metrics_updates() {
    let mut metrics = PerformanceMetrics::default();

    // Simulate trading results
    metrics.sharpe_ratio = 1.8;
    metrics.max_drawdown = 0.05;
    metrics.total_return = 0.25;
    metrics.win_rate = 0.62;
    metrics.trade_count = 150;

    assert!(metrics.sharpe_ratio > 0.0);
    assert!(metrics.win_rate > 0.5);
    assert!(metrics.trade_count > 0);
}

#[tokio::test]
async fn test_metadata_creation() {
    let metadata = ModelMetadata {
        name: "test_model".to_string(),
        model_type: "lstm".to_string(),
        version: "1.0.0".to_string(),
        created_at: chrono::Utc::now(),
        updated_at: chrono::Utc::now(),
        parameters: HashMap::new(),
        input_dimensions: 128,
        description: Some("Test model".to_string()),
    };

    assert_eq!(metadata.name, "test_model");
    assert_eq!(metadata.input_dimensions, 128);
}

#[tokio::test]
async fn test_concentration_metrics() {
    let config = AdaptiveStrategyConfig::default();
    let risk_manager = RiskManager::new(config.risk).unwrap();

    let concentration = risk_manager.get_concentration_metrics().await;
    // May be None if Kelly sizer not initialized
    assert!(concentration.is_ok());
}