// Helper functions for updating persistent learning data
import fs from 'fs/promises';
import path from 'path';

const PERSISTENT_DATA_FILE = path.join(process.cwd(), 'data', 'learning-persistent.json');

async function loadPersistentData() {
  try {
    const data = await fs.readFile(PERSISTENT_DATA_FILE, 'utf8');
    return JSON.parse(data);
  } catch (error) {
    // Return default structure if file doesn't exist
    return {
      totalTrades: 0,
      winningTrades: 0,
      losingTrades: 0,
      totalPnL: 0,
      winRate: 0,
      avgWinAmount: 0,
      avgLossAmount: 0,
      bestTrade: 0,
      worstTrade: 0,
      learningDecisions: 0,
      aiEnhancements: 0,
      riskThresholds: {
        emergency: 1,
        risk: 2,
        mediumRisk: 5
      },
      lastUpdated: new Date().toISOString(),
      systemStatus: 'learning',
      dataCollected: true
    };
  }
}

async function savePersistentData(data) {
  try {
    await fs.writeFile(PERSISTENT_DATA_FILE, JSON.stringify(data, null, 2));
    return true;
  } catch (error) {
    console.error('Error saving persistent data:', error);
    return false;
  }
}

export async function updateTradingStats(tradeData) {
  try {
    const persistentData = await loadPersistentData();
    
    // Update trade counts
    persistentData.totalTrades += 1;
    
    // Determine if trade was winning or losing
    const isWin = tradeData.pnl > 0;
    if (isWin) {
      persistentData.winningTrades += 1;
    } else {
      persistentData.losingTrades += 1;
    }
    
    // Update PnL
    persistentData.totalPnL += tradeData.pnl;
    
    // Update best/worst trades
    if (tradeData.pnl > persistentData.bestTrade) {
      persistentData.bestTrade = tradeData.pnl;
    }
    if (tradeData.pnl < persistentData.worstTrade) {
      persistentData.worstTrade = tradeData.pnl;
    }
    
    // Recalculate averages and win rate
    persistentData.winRate = (persistentData.winningTrades / persistentData.totalTrades) * 100;
    
    if (persistentData.winningTrades > 0) {
      // Calculate average win amount (we need to track this separately for accuracy)
      const winTrades = persistentData.winTrades || [];
      winTrades.push(isWin ? tradeData.pnl : null);
      const wins = winTrades.filter(t => t !== null && t > 0);
      persistentData.avgWinAmount = wins.reduce((sum, win) => sum + win, 0) / wins.length;
    }
    
    if (persistentData.losingTrades > 0) {
      // Calculate average loss amount
      const lossTrades = persistentData.lossTrades || [];
      lossTrades.push(!isWin ? tradeData.pnl : null);
      const losses = lossTrades.filter(t => t !== null && t < 0);
      persistentData.avgLossAmount = losses.reduce((sum, loss) => sum + loss, 0) / losses.length;
    }
    
    // Update timestamp
    persistentData.lastUpdated = new Date().toISOString();
    persistentData.systemStatus = 'active';
    
    // Save updated data
    await savePersistentData(persistentData);
    
    console.log(`📊 Persistent data updated: Trade PnL ${tradeData.pnl}, Total: ${persistentData.totalTrades} trades, ${persistentData.winRate.toFixed(1)}% win rate`);
    
    return persistentData;
  } catch (error) {
    console.error('Error updating trading stats:', error);
    return null;
  }
}

export async function updateLearningDecision() {
  try {
    const persistentData = await loadPersistentData();
    persistentData.learningDecisions += 1;
    persistentData.lastUpdated = new Date().toISOString();
    await savePersistentData(persistentData);
    return persistentData;
  } catch (error) {
    console.error('Error updating learning decision count:', error);
    return null;
  }
}

export async function updateAIEnhancement() {
  try {
    const persistentData = await loadPersistentData();
    persistentData.aiEnhancements += 1;
    persistentData.lastUpdated = new Date().toISOString();
    await savePersistentData(persistentData);
    return persistentData;
  } catch (error) {
    console.error('Error updating AI enhancement count:', error);
    return null;
  }
}