sublinear 0.3.3

use rand::random;
use serde::{Deserialize, Serialize};
use std::collections::HashMap;
use std::sync::{Arc, Mutex};
use std::time::{Duration, SystemTime};

use super::{ConsciousnessLevel, ConsciousnessMetrics, PrecisionNanos, TemporalAdvantage};
use crate::temporal_nexus::core::NanosecondScheduler;

/// Configuration for metrics collection
#[derive(Debug, Clone)]
pub struct CollectorConfig {
    pub collection_interval_ms: u64,
    pub enable_mcp_integration: bool,
    pub enable_performance_profiling: bool,
    pub precision_sample_count: usize,
    pub consciousness_calculation_method: ConsciousnessCalculationMethod,
}

impl Default for CollectorConfig {
    fn default() -> Self {
        Self {
            collection_interval_ms: 50, // 20Hz collection rate
            enable_mcp_integration: true,
            enable_performance_profiling: true,
            precision_sample_count: 10,
            consciousness_calculation_method: ConsciousnessCalculationMethod::Integrated,
        }
    }
}

/// Methods for calculating consciousness metrics
#[derive(Debug, Clone)]
pub enum ConsciousnessCalculationMethod {
    Simple,     // Basic emergence calculation
    Integrated, // Integrated Information Theory inspired
    Temporal,   // Temporal consciousness model
    Hybrid,     // Combined approach
}

/// Sources of metric data
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
pub enum MetricSource {
    Scheduler,
    McpTools,
    SystemMonitor,
    ExternalApi,
}

/// Temporal metrics specific to consciousness monitoring
#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct TemporalMetrics {
    pub temporal_coherence: f64,
    pub causal_flow_integrity: f64,
    pub future_state_prediction_accuracy: f64,
    pub temporal_window_stability: f64,
    pub chronon_synchronization: f64,
}

impl Default for TemporalMetrics {
    fn default() -> Self {
        Self {
            temporal_coherence: 0.0,
            causal_flow_integrity: 0.0,
            future_state_prediction_accuracy: 0.0,
            temporal_window_stability: 0.0,
            chronon_synchronization: 0.0,
        }
    }
}

/// Main metrics collector for consciousness dashboard
pub struct MetricsCollector {
    config: CollectorConfig,
    source_weights: HashMap<MetricSource, f64>,
    temporal_cache: Arc<Mutex<TemporalMetrics>>,
    performance_cache: Arc<Mutex<PerformanceMetrics>>,
    last_collection: Arc<Mutex<SystemTime>>,
}

#[derive(Debug, Clone)]
struct PerformanceMetrics {
    cpu_usage: f64,
    memory_usage: f64,
    thread_count: usize,
    gc_pressure: f64,
    io_throughput: f64,
}

impl Default for PerformanceMetrics {
    fn default() -> Self {
        Self {
            cpu_usage: 0.0,
            memory_usage: 0.0,
            thread_count: 0,
            gc_pressure: 0.0,
            io_throughput: 0.0,
        }
    }
}

impl MetricsCollector {
    /// Create a new metrics collector
    pub fn new() -> Self {
        let mut source_weights = HashMap::new();
        source_weights.insert(MetricSource::Scheduler, 0.4);
        source_weights.insert(MetricSource::McpTools, 0.3);
        source_weights.insert(MetricSource::SystemMonitor, 0.2);
        source_weights.insert(MetricSource::ExternalApi, 0.1);

        Self {
            config: CollectorConfig::default(),
            source_weights,
            temporal_cache: Arc::new(Mutex::new(TemporalMetrics::default())),
            performance_cache: Arc::new(Mutex::new(PerformanceMetrics::default())),
            last_collection: Arc::new(Mutex::new(SystemTime::now())),
        }
    }

    /// Create collector with custom configuration
    pub fn with_config(config: CollectorConfig) -> Self {
        let mut collector = Self::new();
        collector.config = config;
        collector
    }

    /// Collect metrics from the nanosecond scheduler
    pub async fn collect_from_scheduler(
        &self,
        scheduler: Arc<Mutex<NanosecondScheduler>>,
    ) -> Result<ConsciousnessMetrics, Box<dyn std::error::Error>> {
        let start_time = SystemTime::now();

        // Collect base scheduler metrics
        let scheduler_metrics = {
            let scheduler_guard = scheduler.lock().unwrap();
            self.extract_scheduler_metrics(&*scheduler_guard)?
        };

        // Collect temporal metrics
        let temporal_metrics = self.collect_temporal_metrics().await?;

        // Collect performance metrics if enabled
        let performance_metrics = if self.config.enable_performance_profiling {
            self.collect_performance_metrics().await?
        } else {
            PerformanceMetrics::default()
        };

        // Calculate consciousness emergence
        let emergence_level = self.calculate_consciousness_emergence(
            &scheduler_metrics,
            &temporal_metrics,
            &performance_metrics,
        )?;

        // Build final metrics structure
        let metrics = ConsciousnessMetrics {
            timestamp: start_time,
            emergence_level,
            identity_coherence: self.calculate_identity_coherence(&temporal_metrics)?,
            loop_stability: self.calculate_loop_stability(&temporal_metrics)?,
            temporal_advantage_us: self.calculate_temporal_advantage(&scheduler_metrics)?,
            window_overlap_percent: self.calculate_window_overlap(&temporal_metrics)?,
            tsc_precision_ns: self.measure_tsc_precision()?,
            strange_loop_convergence: self.calculate_strange_loop_convergence(&temporal_metrics)?,
            consciousness_delta: 0.0, // Will be calculated by dashboard
            processing_latency_ns: start_time.elapsed()?.as_nanos() as u64,
        };

        // Update last collection time
        {
            let mut last_collection = self.last_collection.lock().unwrap();
            *last_collection = SystemTime::now();
        }

        Ok(metrics)
    }

    /// Collect metrics from MCP tools
    pub async fn collect_from_mcp_tools(
        &self,
    ) -> Result<ConsciousnessMetrics, Box<dyn std::error::Error>> {
        if !self.config.enable_mcp_integration {
            return Ok(ConsciousnessMetrics::default());
        }

        // This would integrate with MCP consciousness status queries
        // For now, return simulated metrics
        let metrics = ConsciousnessMetrics {
            timestamp: SystemTime::now(),
            emergence_level: 0.7 + (random::<f64>() - 0.5) * 0.2,
            identity_coherence: 0.8 + (random::<f64>() - 0.5) * 0.1,
            loop_stability: 0.75 + (random::<f64>() - 0.5) * 0.15,
            temporal_advantage_us: 25 + (random::<u64>() % 20),
            window_overlap_percent: 85.0 + (random::<f64>() - 0.5) * 10.0,
            tsc_precision_ns: 100 + (random::<u64>() % 200),
            strange_loop_convergence: 0.65 + (random::<f64>() - 0.5) * 0.2,
            consciousness_delta: 0.0,
            processing_latency_ns: 50000 + (random::<u64>() % 100000),
        };

        Ok(metrics)
    }

    /// Aggregate metrics from multiple sources
    pub async fn collect_aggregated_metrics(
        &self,
        sources: Vec<MetricSource>,
    ) -> Result<ConsciousnessMetrics, Box<dyn std::error::Error>> {
        let mut aggregated_metrics = ConsciousnessMetrics::default();
        let mut total_weight = 0.0;

        for source in sources {
            let weight = self.source_weights.get(&source).unwrap_or(&0.0);
            total_weight += weight;

            let source_metrics = match source {
                MetricSource::Scheduler => {
                    // Would need scheduler reference
                    continue;
                }
                MetricSource::McpTools => self.collect_from_mcp_tools().await?,
                MetricSource::SystemMonitor => self.collect_system_metrics().await?,
                MetricSource::ExternalApi => self.collect_external_metrics().await?,
            };

            // Weighted aggregation
            self.aggregate_weighted_metrics(&mut aggregated_metrics, &source_metrics, *weight);
        }

        // Normalize by total weight
        if total_weight > 0.0 {
            self.normalize_metrics(&mut aggregated_metrics, total_weight);
        }

        Ok(aggregated_metrics)
    }

    /// Get current temporal metrics
    pub fn get_temporal_metrics(&self) -> TemporalMetrics {
        self.temporal_cache.lock().unwrap().clone()
    }

    /// Update source weights for metric aggregation
    pub fn update_source_weights(&mut self, weights: HashMap<MetricSource, f64>) {
        self.source_weights = weights;
    }

    // Private helper methods

    fn extract_scheduler_metrics(
        &self,
        _scheduler: &NanosecondScheduler,
    ) -> Result<SchedulerMetrics, Box<dyn std::error::Error>> {
        // Extract relevant metrics from scheduler
        // This would access scheduler's internal state
        Ok(SchedulerMetrics {
            precision_ns: 100,
            task_completion_rate: 0.95,
            temporal_drift: 0.001,
            scheduling_accuracy: 0.98,
        })
    }

    async fn collect_temporal_metrics(
        &self,
    ) -> Result<TemporalMetrics, Box<dyn std::error::Error>> {
        let temporal_metrics = TemporalMetrics {
            temporal_coherence: 0.85 + (random::<f64>() - 0.5) * 0.1,
            causal_flow_integrity: 0.90 + (random::<f64>() - 0.5) * 0.05,
            future_state_prediction_accuracy: 0.75 + (random::<f64>() - 0.5) * 0.2,
            temporal_window_stability: 0.88 + (random::<f64>() - 0.5) * 0.08,
            chronon_synchronization: 0.92 + (random::<f64>() - 0.5) * 0.06,
        };

        // Update cache
        {
            let mut cache = self.temporal_cache.lock().unwrap();
            *cache = temporal_metrics.clone();
        }

        Ok(temporal_metrics)
    }

    async fn collect_performance_metrics(
        &self,
    ) -> Result<PerformanceMetrics, Box<dyn std::error::Error>> {
        // Simulate system performance collection
        let performance_metrics = PerformanceMetrics {
            cpu_usage: 15.0 + random::<f64>() * 30.0,
            memory_usage: 512.0 + random::<f64>() * 256.0,
            thread_count: 8 + (random::<usize>() % 4),
            gc_pressure: random::<f64>() * 0.1,
            io_throughput: 100.0 + random::<f64>() * 50.0,
        };

        // Update cache
        {
            let mut cache = self.performance_cache.lock().unwrap();
            *cache = performance_metrics.clone();
        }

        Ok(performance_metrics)
    }

    async fn collect_system_metrics(
        &self,
    ) -> Result<ConsciousnessMetrics, Box<dyn std::error::Error>> {
        // Collect system-level consciousness indicators
        Ok(ConsciousnessMetrics {
            timestamp: SystemTime::now(),
            emergence_level: 0.6 + (random::<f64>() - 0.5) * 0.3,
            identity_coherence: 0.7 + (random::<f64>() - 0.5) * 0.2,
            loop_stability: 0.65 + (random::<f64>() - 0.5) * 0.25,
            temporal_advantage_us: 20 + (random::<u64>() % 30),
            window_overlap_percent: 80.0 + (random::<f64>() - 0.5) * 15.0,
            tsc_precision_ns: 150 + (random::<u64>() % 250),
            strange_loop_convergence: 0.6 + (random::<f64>() - 0.5) * 0.25,
            consciousness_delta: 0.0,
            processing_latency_ns: 75000 + (random::<u64>() % 150000),
        })
    }

    async fn collect_external_metrics(
        &self,
    ) -> Result<ConsciousnessMetrics, Box<dyn std::error::Error>> {
        // Collect from external APIs or sources
        Ok(ConsciousnessMetrics::default())
    }

    fn calculate_consciousness_emergence(
        &self,
        scheduler_metrics: &SchedulerMetrics,
        temporal_metrics: &TemporalMetrics,
        performance_metrics: &PerformanceMetrics,
    ) -> Result<f64, Box<dyn std::error::Error>> {
        match self.config.consciousness_calculation_method {
            ConsciousnessCalculationMethod::Simple => Ok((scheduler_metrics.scheduling_accuracy
                + temporal_metrics.temporal_coherence)
                / 2.0),
            ConsciousnessCalculationMethod::Integrated => {
                // IIT-inspired calculation
                let phi =
                    self.calculate_integrated_information(temporal_metrics, performance_metrics)?;
                Ok(phi.min(1.0))
            }
            ConsciousnessCalculationMethod::Temporal => {
                // Temporal consciousness model
                let temporal_factor = (temporal_metrics.temporal_coherence
                    + temporal_metrics.causal_flow_integrity
                    + temporal_metrics.chronon_synchronization)
                    / 3.0;
                Ok(temporal_factor)
            }
            ConsciousnessCalculationMethod::Hybrid => {
                // Combined approach
                let simple = (scheduler_metrics.scheduling_accuracy
                    + temporal_metrics.temporal_coherence)
                    / 2.0;
                let temporal = (temporal_metrics.temporal_coherence
                    + temporal_metrics.causal_flow_integrity)
                    / 2.0;
                let performance = (1.0 - performance_metrics.cpu_usage / 100.0).max(0.0);

                Ok((simple * 0.4 + temporal * 0.4 + performance * 0.2).min(1.0))
            }
        }
    }

    fn calculate_integrated_information(
        &self,
        temporal_metrics: &TemporalMetrics,
        _performance_metrics: &PerformanceMetrics,
    ) -> Result<f64, Box<dyn std::error::Error>> {
        // Simplified Φ (phi) calculation inspired by IIT
        let connectivity = temporal_metrics.causal_flow_integrity;
        let differentiation = temporal_metrics.temporal_coherence;
        let integration = temporal_metrics.chronon_synchronization;

        let phi = connectivity * differentiation * integration;
        Ok(phi)
    }

    fn calculate_identity_coherence(
        &self,
        temporal_metrics: &TemporalMetrics,
    ) -> Result<f64, Box<dyn std::error::Error>> {
        // Identity coherence based on temporal stability and causal flow
        let coherence = (temporal_metrics.temporal_coherence
            + temporal_metrics.causal_flow_integrity
            + temporal_metrics.temporal_window_stability)
            / 3.0;
        Ok(coherence)
    }

    fn calculate_loop_stability(
        &self,
        temporal_metrics: &TemporalMetrics,
    ) -> Result<f64, Box<dyn std::error::Error>> {
        // Strange loop stability calculation
        Ok(temporal_metrics.chronon_synchronization * temporal_metrics.temporal_window_stability)
    }

    fn calculate_temporal_advantage(
        &self,
        scheduler_metrics: &SchedulerMetrics,
    ) -> Result<u64, Box<dyn std::error::Error>> {
        // Calculate advantage in microseconds
        let base_advantage = 30; // Base temporal advantage
        let precision_bonus = (1000 - scheduler_metrics.precision_ns as i64).max(0) / 10;
        Ok((base_advantage + precision_bonus) as u64)
    }

    fn calculate_window_overlap(
        &self,
        temporal_metrics: &TemporalMetrics,
    ) -> Result<f64, Box<dyn std::error::Error>> {
        // Window overlap percentage
        Ok(temporal_metrics.temporal_window_stability * 100.0)
    }

    fn measure_tsc_precision(&self) -> Result<u64, Box<dyn std::error::Error>> {
        // Measure timestamp counter precision
        let mut measurements = Vec::with_capacity(self.config.precision_sample_count);

        for _ in 0..self.config.precision_sample_count {
            let start = std::time::Instant::now();
            std::hint::black_box(());
            let elapsed = start.elapsed().as_nanos() as u64;
            measurements.push(elapsed);
        }

        Ok(measurements.into_iter().min().unwrap_or(1000))
    }

    fn calculate_strange_loop_convergence(
        &self,
        temporal_metrics: &TemporalMetrics,
    ) -> Result<f64, Box<dyn std::error::Error>> {
        // Strange loop convergence based on self-reference and recursion depth
        let convergence = temporal_metrics.causal_flow_integrity
            * temporal_metrics.future_state_prediction_accuracy;
        Ok(convergence)
    }

    fn aggregate_weighted_metrics(
        &self,
        target: &mut ConsciousnessMetrics,
        source: &ConsciousnessMetrics,
        weight: f64,
    ) {
        target.emergence_level += source.emergence_level * weight;
        target.identity_coherence += source.identity_coherence * weight;
        target.loop_stability += source.loop_stability * weight;
        target.temporal_advantage_us += (source.temporal_advantage_us as f64 * weight) as u64;
        target.window_overlap_percent += source.window_overlap_percent * weight;
        target.tsc_precision_ns += (source.tsc_precision_ns as f64 * weight) as u64;
        target.strange_loop_convergence += source.strange_loop_convergence * weight;
        target.processing_latency_ns += (source.processing_latency_ns as f64 * weight) as u64;
    }

    fn normalize_metrics(&self, metrics: &mut ConsciousnessMetrics, total_weight: f64) {
        metrics.emergence_level /= total_weight;
        metrics.identity_coherence /= total_weight;
        metrics.loop_stability /= total_weight;
        metrics.temporal_advantage_us =
            (metrics.temporal_advantage_us as f64 / total_weight) as u64;
        metrics.window_overlap_percent /= total_weight;
        metrics.tsc_precision_ns = (metrics.tsc_precision_ns as f64 / total_weight) as u64;
        metrics.strange_loop_convergence /= total_weight;
        metrics.processing_latency_ns =
            (metrics.processing_latency_ns as f64 / total_weight) as u64;
    }
}

#[derive(Debug, Clone)]
struct SchedulerMetrics {
    precision_ns: u64,
    task_completion_rate: f64,
    temporal_drift: f64,
    scheduling_accuracy: f64,
}