<?php

declare(strict_types=1);

/**
 * ML Monitoring Dashboard Data Collection Test
 *
 * Demonstrates comprehensive monitoring data collection for ML systems:
 * 1. Model performance metrics (accuracy, precision, recall, F1)
 * 2. Prediction distribution and confidence histograms
 * 3. Model version comparison and drift detection
 * 4. A/B test status and progress
 * 5. System health indicators
 * 6. Performance degradation alerts
 */

require_once __DIR__ . '/../../vendor/autoload.php';

use App\Framework\MachineLearning\ModelManagement\InMemoryModelRegistry;
use App\Framework\MachineLearning\ModelManagement\ModelPerformanceMonitor;
use App\Framework\MachineLearning\ModelManagement\InMemoryPerformanceStorage;
use App\Framework\MachineLearning\ModelManagement\NullAlertingService;
use App\Framework\MachineLearning\ModelManagement\ABTestingService;
use App\Framework\MachineLearning\ModelManagement\AutoTuningEngine;
use App\Framework\MachineLearning\ModelManagement\ValueObjects\ModelMetadata;
use App\Framework\MachineLearning\ModelManagement\ValueObjects\ModelType;
use App\Framework\MachineLearning\ModelManagement\ValueObjects\ABTestConfig;
use App\Framework\Queue\MachineLearning\QueueAnomalyModelAdapter;
use App\Framework\Queue\MachineLearning\JobAnomalyDetector;
use App\Framework\Waf\MachineLearning\WafBehavioralModelAdapter;
use App\Framework\Waf\MachineLearning\BehaviorAnomalyDetector;
use App\Framework\Core\ValueObjects\Version;
use App\Framework\Core\ValueObjects\Timestamp;
use App\Framework\Core\ValueObjects\Duration;
use App\Framework\Core\ValueObjects\Score;
use App\Framework\Random\SecureRandomGenerator;

echo "=== ML Monitoring Dashboard Data Collection Test ===\n\n";

try {
    // ========================================================================
    // Setup: Initialize infrastructure
    // ========================================================================
    echo "1. Initializing ML Monitoring Infrastructure...\n";

    $registry = new InMemoryModelRegistry();
    $storage = new InMemoryPerformanceStorage();
    $alerting = new NullAlertingService();
    $performanceMonitor = new ModelPerformanceMonitor($registry, $storage, $alerting);
    $random = new SecureRandomGenerator();
    $abTesting = new ABTestingService($random, $registry);
    $autoTuning = new AutoTuningEngine($performanceMonitor, $registry, $storage);

    echo "   ✓ Infrastructure initialized\n\n";

    // ========================================================================
    // Setup: Register multiple models with different types
    // ========================================================================
    echo "2. Registering multiple ML models...\n";

    $models = [
        'queue-anomaly' => [
            'type' => ModelType::UNSUPERVISED,
            'version' => Version::fromString('1.0.0'),
            'config' => ['threshold' => 0.4, 'z_score_threshold' => 3.0]
        ],
        'waf-behavioral' => [
            'type' => ModelType::UNSUPERVISED,
            'version' => Version::fromString('1.2.0'),
            'config' => ['threshold' => 0.5, 'z_score_threshold' => 2.5]
        ],
        'fraud-detector' => [
            'type' => ModelType::SUPERVISED,
            'version' => Version::fromString('2.0.0'),
            'config' => ['threshold' => 0.7, 'algorithm' => 'xgboost']
        ],
        'spam-classifier' => [
            'type' => ModelType::SUPERVISED,
            'version' => Version::fromString('1.5.0'),
            'config' => ['threshold' => 0.6, 'algorithm' => 'naive_bayes']
        ],
    ];

    foreach ($models as $modelName => $info) {
        $metadata = new ModelMetadata(
            modelName: $modelName,
            modelType: $info['type'],
            version: $info['version'],
            configuration: $info['config'],
            createdAt: Timestamp::now()
        );
        $registry->register($metadata);
        echo "   ✓ Registered: {$modelName} v{$info['version']->toString()} ({$info['type']->value})\n";
    }
    echo "\n";

    // ========================================================================
    // Setup: Simulate prediction data for all models
    // ========================================================================
    echo "3. Simulating prediction data...\n";

    $timestamp = Timestamp::now();

    // Queue Anomaly: 95% accuracy
    $queuePredictions = [
        ...array_fill(0, 95, ['confidence' => 0.85, 'actual' => true, 'prediction' => true]),
        ...array_fill(0, 5, ['confidence' => 0.45, 'actual' => false, 'prediction' => true]),
    ];

    foreach ($queuePredictions as $pred) {
        $storage->storePrediction([
            'model_name' => 'queue-anomaly',
            'version' => '1.0.0',
            'prediction' => $pred['prediction'],
            'actual' => $pred['actual'],
            'confidence' => $pred['confidence'],
            'features' => [],
            'timestamp' => $timestamp->toDateTime(),
            'is_correct' => $pred['prediction'] === $pred['actual'],
        ]);
    }

    // WAF Behavioral: 88% accuracy
    $wafPredictions = [
        ...array_fill(0, 88, ['confidence' => 0.9, 'actual' => true, 'prediction' => true]),
        ...array_fill(0, 12, ['confidence' => 0.55, 'actual' => false, 'prediction' => true]),
    ];

    foreach ($wafPredictions as $pred) {
        $storage->storePrediction([
            'model_name' => 'waf-behavioral',
            'version' => '1.2.0',
            'prediction' => $pred['prediction'],
            'actual' => $pred['actual'],
            'confidence' => $pred['confidence'],
            'features' => [],
            'timestamp' => $timestamp->toDateTime(),
            'is_correct' => $pred['prediction'] === $pred['actual'],
        ]);
    }

    // Fraud Detector: 92% accuracy
    $fraudPredictions = [
        ...array_fill(0, 92, ['confidence' => 0.95, 'actual' => true, 'prediction' => true]),
        ...array_fill(0, 8, ['confidence' => 0.6, 'actual' => false, 'prediction' => true]),
    ];

    foreach ($fraudPredictions as $pred) {
        $storage->storePrediction([
            'model_name' => 'fraud-detector',
            'version' => '2.0.0',
            'prediction' => $pred['prediction'],
            'actual' => $pred['actual'],
            'confidence' => $pred['confidence'],
            'features' => [],
            'timestamp' => $timestamp->toDateTime(),
            'is_correct' => $pred['prediction'] === $pred['actual'],
        ]);
    }

    // Spam Classifier: 78% accuracy (degraded)
    $spamPredictions = [
        ...array_fill(0, 78, ['confidence' => 0.7, 'actual' => true, 'prediction' => true]),
        ...array_fill(0, 22, ['confidence' => 0.65, 'actual' => false, 'prediction' => true]),
    ];

    foreach ($spamPredictions as $pred) {
        $storage->storePrediction([
            'model_name' => 'spam-classifier',
            'version' => '1.5.0',
            'prediction' => $pred['prediction'],
            'actual' => $pred['actual'],
            'confidence' => $pred['confidence'],
            'features' => [],
            'timestamp' => $timestamp->toDateTime(),
            'is_correct' => $pred['prediction'] === $pred['actual'],
        ]);
    }

    echo "   ✓ Simulated 400 total predictions across 4 models\n\n";

    // ========================================================================
    // Dashboard Data 1: Model Performance Overview
    // ========================================================================
    echo "4. Collecting Model Performance Overview...\n";

    $performanceOverview = [];

    foreach (array_keys($models) as $modelName) {
        $metadata = $registry->get($modelName, $models[$modelName]['version']);
        $metrics = $performanceMonitor->getCurrentMetrics($modelName, $models[$modelName]['version']);

        $performanceOverview[$modelName] = [
            'version' => $models[$modelName]['version']->toString(),
            'type' => $models[$modelName]['type']->value,
            'accuracy' => $metrics['accuracy'],
            'precision' => $metrics['precision'] ?? 0.0,
            'recall' => $metrics['recall'] ?? 0.0,
            'f1_score' => $metrics['f1_score'] ?? 0.0,
            'total_predictions' => $metrics['total_predictions'],
            'average_confidence' => $metrics['average_confidence'] ?? 0.0,
            'threshold' => $models[$modelName]['config']['threshold'],
            'status' => $metrics['accuracy'] >= 0.85 ? 'healthy' : 'degraded'
        ];
    }

    echo "   → Performance Overview:\n";
    foreach ($performanceOverview as $modelName => $data) {
        echo "      {$modelName}:\n";
        echo "         Accuracy: " . sprintf("%.1f%%", $data['accuracy'] * 100) . "\n";
        echo "         Precision: " . sprintf("%.1f%%", $data['precision'] * 100) . "\n";
        echo "         Recall: " . sprintf("%.1f%%", $data['recall'] * 100) . "\n";
        echo "         F1-Score: " . sprintf("%.1f%%", $data['f1_score'] * 100) . "\n";
        echo "         Predictions: {$data['total_predictions']}\n";
        echo "         Status: {$data['status']}\n";
    }
    echo "\n";

    // ========================================================================
    // Dashboard Data 2: Performance Degradation Alerts
    // ========================================================================
    echo "5. Checking Performance Degradation Alerts...\n";

    $degradationAlerts = [];

    foreach (array_keys($models) as $modelName) {
        $metrics = $performanceMonitor->getCurrentMetrics($modelName, $models[$modelName]['version']);

        if ($metrics['accuracy'] < 0.85) {
            $degradationAlerts[] = [
                'model_name' => $modelName,
                'version' => $models[$modelName]['version']->toString(),
                'current_accuracy' => $metrics['accuracy'],
                'threshold' => 0.85,
                'severity' => $metrics['accuracy'] < 0.7 ? 'critical' : 'warning',
                'recommendation' => 'Consider retraining or rolling back to previous version'
            ];
        }
    }

    echo "   → Degradation Alerts: " . count($degradationAlerts) . " alert(s)\n";
    foreach ($degradationAlerts as $alert) {
        echo "      [{$alert['severity']}] {$alert['model_name']} v{$alert['version']}\n";
        echo "         Accuracy: " . sprintf("%.1f%%", $alert['current_accuracy'] * 100) . " (threshold: " . sprintf("%.0f%%", $alert['threshold'] * 100) . ")\n";
        echo "         Recommendation: {$alert['recommendation']}\n";
    }
    echo "\n";

    // ========================================================================
    // Dashboard Data 3: Confusion Matrix Breakdown
    // ========================================================================
    echo "6. Collecting Confusion Matrix Data...\n";

    $confusionMatrices = [];

    foreach (array_keys($models) as $modelName) {
        $metrics = $performanceMonitor->getCurrentMetrics($modelName, $models[$modelName]['version']);

        if (isset($metrics['confusion_matrix'])) {
            $confusionMatrices[$modelName] = [
                'true_positive' => $metrics['confusion_matrix']['true_positive'],
                'true_negative' => $metrics['confusion_matrix']['true_negative'],
                'false_positive' => $metrics['confusion_matrix']['false_positive'],
                'false_negative' => $metrics['confusion_matrix']['false_negative'],
                'total' => $metrics['total_predictions'],
                'false_positive_rate' => $metrics['confusion_matrix']['false_positive'] / $metrics['total_predictions'],
                'false_negative_rate' => $metrics['confusion_matrix']['false_negative'] / $metrics['total_predictions'],
            ];
        }
    }

    echo "   → Confusion Matrices:\n";
    foreach ($confusionMatrices as $modelName => $matrix) {
        echo "      {$modelName}:\n";
        echo "         TP: {$matrix['true_positive']}, TN: {$matrix['true_negative']}\n";
        echo "         FP: {$matrix['false_positive']}, FN: {$matrix['false_negative']}\n";
        echo "         FP Rate: " . sprintf("%.1f%%", $matrix['false_positive_rate'] * 100) . "\n";
        echo "         FN Rate: " . sprintf("%.1f%%", $matrix['false_negative_rate'] * 100) . "\n";
    }
    echo "\n";

    // ========================================================================
    // Dashboard Data 4: Model Registry Summary
    // ========================================================================
    echo "7. Collecting Model Registry Summary...\n";

    $registrySummary = [
        'total_models' => $registry->getTotalCount(),
        'total_model_types' => count($registry->getAllModelNames()),
        'models_by_type' => [
            'supervised' => 0,
            'unsupervised' => 0,
            'reinforcement' => 0
        ],
        'average_predictions_per_model' => 0
    ];

    $totalPredictions = 0;

    foreach (array_keys($models) as $modelName) {
        $metadata = $registry->get($modelName, $models[$modelName]['version']);
        $registrySummary['models_by_type'][$metadata->modelType->value]++;

        $metrics = $performanceMonitor->getCurrentMetrics($modelName, $models[$modelName]['version']);
        $totalPredictions += $metrics['total_predictions'];
    }

    $registrySummary['average_predictions_per_model'] = $totalPredictions / $registrySummary['total_model_types'];

    echo "   → Registry Summary:\n";
    echo "      Total Models: {$registrySummary['total_models']}\n";
    echo "      Model Types: {$registrySummary['total_model_types']}\n";
    echo "      Supervised: {$registrySummary['models_by_type']['supervised']}\n";
    echo "      Unsupervised: {$registrySummary['models_by_type']['unsupervised']}\n";
    echo "      Avg Predictions/Model: " . sprintf("%.0f", $registrySummary['average_predictions_per_model']) . "\n\n";

    // ========================================================================
    // Dashboard Data 5: System Health Indicators
    // ========================================================================
    echo "8. Collecting System Health Indicators...\n";

    $healthIndicators = [
        'overall_status' => 'healthy',
        'healthy_models' => 0,
        'degraded_models' => 0,
        'average_accuracy' => 0.0,
        'lowest_accuracy' => 1.0,
        'highest_accuracy' => 0.0,
        'total_predictions' => $totalPredictions,
        'models_below_threshold' => []
    ];

    $totalAccuracy = 0.0;

    foreach (array_keys($models) as $modelName) {
        $metrics = $performanceMonitor->getCurrentMetrics($modelName, $models[$modelName]['version']);

        if ($metrics['accuracy'] >= 0.85) {
            $healthIndicators['healthy_models']++;
        } else {
            $healthIndicators['degraded_models']++;
            $healthIndicators['models_below_threshold'][] = $modelName;
        }

        $totalAccuracy += $metrics['accuracy'];

        if ($metrics['accuracy'] < $healthIndicators['lowest_accuracy']) {
            $healthIndicators['lowest_accuracy'] = $metrics['accuracy'];
        }

        if ($metrics['accuracy'] > $healthIndicators['highest_accuracy']) {
            $healthIndicators['highest_accuracy'] = $metrics['accuracy'];
        }
    }

    $healthIndicators['average_accuracy'] = $totalAccuracy / count($models);

    if ($healthIndicators['degraded_models'] > 0) {
        $healthIndicators['overall_status'] = $healthIndicators['degraded_models'] > 2 ? 'critical' : 'warning';
    }

    echo "   → Health Indicators:\n";
    echo "      Overall Status: {$healthIndicators['overall_status']}\n";
    echo "      Healthy Models: {$healthIndicators['healthy_models']}/{" . count($models) . "}\n";
    echo "      Degraded Models: {$healthIndicators['degraded_models']}\n";
    echo "      Average Accuracy: " . sprintf("%.1f%%", $healthIndicators['average_accuracy'] * 100) . "\n";
    echo "      Accuracy Range: " . sprintf("%.1f%%", $healthIndicators['lowest_accuracy'] * 100) . " - " . sprintf("%.1f%%", $healthIndicators['highest_accuracy'] * 100) . "\n";
    echo "      Total Predictions: {$healthIndicators['total_predictions']}\n";

    if (!empty($healthIndicators['models_below_threshold'])) {
        echo "      Models Below Threshold: " . implode(', ', $healthIndicators['models_below_threshold']) . "\n";
    }
    echo "\n";

    // ========================================================================
    // Dashboard Data 6: JSON Export for Frontend
    // ========================================================================
    echo "9. Generating JSON Dashboard Data...\n";

    $dashboardData = [
        'timestamp' => Timestamp::now()->format('Y-m-d H:i:s'),
        'summary' => [
            'total_models' => $registrySummary['total_models'],
            'healthy_models' => $healthIndicators['healthy_models'],
            'degraded_models' => $healthIndicators['degraded_models'],
            'total_predictions' => $healthIndicators['total_predictions'],
            'average_accuracy' => $healthIndicators['average_accuracy'],
            'overall_status' => $healthIndicators['overall_status']
        ],
        'models' => $performanceOverview,
        'alerts' => $degradationAlerts,
        'confusion_matrices' => $confusionMatrices,
        'health' => $healthIndicators
    ];

    $jsonData = json_encode($dashboardData, JSON_PRETTY_PRINT);

    echo "   ✓ JSON Dashboard Data Generated (" . strlen($jsonData) . " bytes)\n";
    echo "\n";

    // ========================================================================
    // Display JSON Sample
    // ========================================================================
    echo "10. Dashboard Data Sample (JSON):\n";
    echo substr($jsonData, 0, 500) . "...\n\n";

    // ========================================================================
    // Test Summary
    // ========================================================================
    echo "=== Test Summary ===\n";
    echo "✓ Model Performance Overview: Collected\n";
    echo "✓ Degradation Alerts: Generated\n";
    echo "✓ Confusion Matrices: Calculated\n";
    echo "✓ Registry Summary: Compiled\n";
    echo "✓ System Health Indicators: Analyzed\n";
    echo "✓ JSON Dashboard Data: Exported\n\n";

    echo "Dashboard Summary:\n";
    echo "  - {$registrySummary['total_models']} models tracked\n";
    echo "  - {$healthIndicators['healthy_models']} healthy, {$healthIndicators['degraded_models']} degraded\n";
    echo "  - Average accuracy: " . sprintf("%.1f%%", $healthIndicators['average_accuracy'] * 100) . "\n";
    echo "  - {$totalPredictions} total predictions processed\n";
    echo "  - " . count($degradationAlerts) . " active alert(s)\n";
    echo "  - Overall status: {$healthIndicators['overall_status']}\n\n";

    echo "=== ML Monitoring Dashboard PASSED ===\n";

} catch (\Throwable $e) {
    echo "\n!!! TEST FAILED !!!\n";
    echo "Error: " . $e->getMessage() . "\n";
    echo "File: " . $e->getFile() . ":" . $e->getLine() . "\n";
    echo "\nStack trace:\n" . $e->getTraceAsString() . "\n";
    exit(1);
}