tools/CryptoAnalyzer.py

import ccxt
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.dates as mdates
from datetime import datetime, timedelta
import ta
from typing import List, Dict, Any, Optional, Tuple, Union, Callable
import os
import logging
import json
import time
from dataclasses import dataclass, asdict
import traceback

# Configure logging
logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
logger = logging.getLogger(__name__)

@dataclass
class AnalysisResult:
    """Data class for structured analysis results"""
    ticker: str
    timeframe: str
    summary: Dict[str, Any]
    chart_path: str
    indicators_used: List[str]
    data_points: int
    period: str
    timestamp: str
    
    def to_dict(self) -> Dict[str, Any]:
        """Convert to dictionary"""
        return asdict(self)
    
    def to_json(self) -> str:
        """Convert to JSON string"""
        return json.dumps(self.to_dict(), indent=2)
    
    def get_trading_signal(self) -> Tuple[str, float]:
        """Extract trading signal from analysis"""
        signal = "NEUTRAL"
        confidence = 0.5
        
        # Extract trend info
        if 'trend' in self.summary:
            if self.summary['trend'] == "Bullish":
                signal = "BUY"
                confidence = 0.7
            elif self.summary['trend'] == "Bearish":
                signal = "SELL"
                confidence = 0.7
        
        # Factor in RSI
        if 'RSI' in self.summary:
            rsi_value = self.summary['RSI']['value']
            if rsi_value < 30 and signal != "BUY":
                signal = "BUY"
                confidence = max(confidence, 0.8)
            elif rsi_value > 70 and signal != "SELL":
                signal = "SELL"
                confidence = max(confidence, 0.8)
        
        # Consider MACD
        if 'MACD' in self.summary:
            if self.summary['MACD']['interpretation'] == "Bullish crossover" and signal != "SELL":
                signal = "BUY"
                confidence = max(confidence, 0.75)
            elif self.summary['MACD']['interpretation'] == "Bearish crossover" and signal != "BUY":
                signal = "SELL"
                confidence = max(confidence, 0.75)
        
        return signal, confidence
    
    def get_summary_text(self) -> str:
        """Get summary as formatted text"""
        summary_lines = [f"**Analysis for {self.ticker} ({self.timeframe}):**\n"]
        
        # Add trend information
        if 'trend' in self.summary:
            summary_lines.append(f"Overall Trend: {self.summary['trend']}")
        
        # Add price information
        if 'price' in self.summary:
            price_info = self.summary['price']
            change_text = f" (24h change: {price_info['change_24h']}%)" if price_info['change_24h'] is not None else ""
            summary_lines.append(f"Current Price: {price_info['current']}{change_text}")
        
        # Add RSI information
        if 'RSI' in self.summary:
            rsi_info = self.summary['RSI']
            summary_lines.append(f"RSI: {rsi_info['value']} - {rsi_info['interpretation']}")
        
        # Add MACD information
        if 'MACD' in self.summary:
            macd_info = self.summary['MACD']
            summary_lines.append(f"MACD: {macd_info['value']}, Signal: {macd_info['signal']}")
            summary_lines.append(f"MACD Interpretation: {macd_info['interpretation']}")
        
        # Add Bollinger Bands information
        if 'Bollinger_Bands' in self.summary:
            bb_info = self.summary['Bollinger_Bands']
            summary_lines.append(f"Bollinger Bands: Upper: {bb_info['upper']}, Middle: {bb_info['middle']}, Lower: {bb_info['lower']}")
            summary_lines.append(f"Bandwidth: {bb_info['bandwidth']}%, Position: {bb_info['position']}, Squeeze: {bb_info['squeeze']}")
        
        # Add support and resistance levels if available
        if 'Support' in self.summary and 'Resistance' in self.summary:
            summary_lines.append(f"Support Levels: {self.summary['Support']}")
            summary_lines.append(f"Resistance Levels: {self.summary['Resistance']}")
        
        # Add trading signal
        signal, confidence = self.get_trading_signal()
        summary_lines.append(f"\nTrading Signal: {signal} (Confidence: {confidence:.2f})")
        
        # Add chart path
        summary_lines.append(f"\nChart saved to: {self.chart_path}")
        
        # Add analysis period
        summary_lines.append(f"Analysis period: {self.period}")
        
        return "\n".join(summary_lines)

class CryptoAnalyzer:
    """A class to analyze cryptocurrency charts with technical indicators."""
    
    def __init__(self, 
                exchange_name: str = "binance", 
                output_dir: str = "./charts",
                rate_limit_pause: float = 1.0,
                max_retries: int = 3):
        """
        Initialize the crypto analyzer with an exchange.
        
        Args:
            exchange_name (str): Name of CCXT-supported exchange (default: "binance")
            output_dir (str): Directory to save chart images
            rate_limit_pause (float): Pause between API calls to avoid rate limits
            max_retries (int): Maximum number of API call retries
        """
        try:
            self.exchange = getattr(ccxt, exchange_name)()
            self.output_dir = output_dir
            self.rate_limit_pause = rate_limit_pause
            self.max_retries = max_retries
            self.supports_advanced_patterns = True  # Flag for pattern recognition capabilities
            
            # Create output directory if it doesn't exist
            if not os.path.exists(output_dir):
                os.makedirs(output_dir)
                
            # Cache for market data to reduce API calls
            self._market_cache = {}
            self._last_api_call = 0
            
        except AttributeError:
            supported = ", ".join(ccxt.exchanges)
            raise ValueError(f"Exchange '{exchange_name}' not supported. Choose from: {supported}")
        except Exception as e:
            raise Exception(f"Failed to initialize analyzer: {str(e)}")
    
    def get_supported_exchanges(self) -> List[str]:
        """Return list of supported exchanges"""
        return ccxt.exchanges
    
    def get_supported_timeframes(self) -> List[str]:
        """Return list of supported timeframes for current exchange"""
        return list(self.exchange.timeframes.keys()) if hasattr(self.exchange, 'timeframes') else ["1m", "5m", "15m", "30m", "1h", "4h", "1d", "1w"]
    
    def get_supported_indicators(self) -> Dict[str, str]:
        """Return dictionary of supported indicators with descriptions"""
        return {
            "RSI": "Relative Strength Index - Momentum oscillator measuring speed and change of price movements",
            "MACD": "Moving Average Convergence Divergence - Trend-following momentum indicator",
            "SMA": "Simple Moving Average - Average price over specified period",
            "EMA": "Exponential Moving Average - Weighted moving average giving more importance to recent prices",
            "BB": "Bollinger Bands - Volatility indicator showing price channels around moving average",
            "ATR": "Average True Range - Volatility indicator measuring market volatility",
            "OBV": "On-Balance Volume - Volume indicator using volume flow to predict changes in price",
            "VWAP": "Volume Weighted Average Price - Average price weighted by volume",
            "Ichimoku": "Ichimoku Cloud - Trend indicator showing support/resistance levels and momentum",
            "Stochastic": "Stochastic Oscillator - Momentum indicator comparing close price to price range",
            "Patterns": "Candlestick pattern recognition for common bullish and bearish patterns"
        }
    
    def _respect_rate_limit(self):
        """Implement rate limiting to avoid API restrictions"""
        elapsed = time.time() - self._last_api_call
        if elapsed < self.rate_limit_pause:
            time.sleep(self.rate_limit_pause - elapsed)
        self._last_api_call = time.time()
    
    def get_available_pairs(self, quote_currency: Optional[str] = None) -> List[str]:
        """
        Get available trading pairs from exchange
        
        Args:
            quote_currency (Optional[str]): Filter by quote currency (e.g., "USD", "BTC")
            
        Returns:
            List[str]: List of available trading pairs
        """
        try:
            if 'markets' not in self._market_cache:
                self._respect_rate_limit()
                self._market_cache['markets'] = self.exchange.load_markets()
            
            pairs = list(self._market_cache['markets'].keys())
            
            if quote_currency:
                pairs = [p for p in pairs if p.endswith(f"/{quote_currency}")]
                
            return pairs
        except Exception as e:
            logger.error(f"Error fetching available pairs: {str(e)}")
            return []
    
    def fetch_data(self, 
                  ticker: str, 
                  timeframe: str, 
                  days: int = 30, 
                  retry_on_error: bool = True) -> pd.DataFrame:
        """
        Fetch OHLCV data for a specified ticker and timeframe.
        
        Args:
            ticker (str): Trading pair (e.g., "BTC/USD")
            timeframe (str): Timeframe (e.g., "1h", "4h", "1d")
            days (int): Number of days of historical data to fetch
            retry_on_error (bool): Whether to retry on network errors
            
        Returns:
            pd.DataFrame: DataFrame with OHLCV data
        """
        retries = 0
        last_error = None
        
        while retries <= self.max_retries:
            try:
                # Format symbol according to exchange requirements
                symbol = ticker
                
                # Calculate timestamp for the specified number of days ago
                since = int((datetime.now() - timedelta(days=days)).timestamp() * 1000)
                
                # Fetch OHLCV data with rate limiting
                self._respect_rate_limit()
                logger.info(f"Fetching {days} days of {timeframe} data for {ticker}")
                ohlcv = self.exchange.fetch_ohlcv(symbol, timeframe=timeframe, since=since, limit=1000)
                
                # Check if we got data
                if not ohlcv or len(ohlcv) < 2:
                    raise ValueError(f"Insufficient data returned for {ticker} ({timeframe})")
                
                # Convert to DataFrame
                df = pd.DataFrame(ohlcv, columns=['timestamp', 'open', 'high', 'low', 'close', 'volume'])
                df['timestamp'] = pd.to_datetime(df['timestamp'], unit='ms')
                df.set_index('timestamp', inplace=True)
                
                return df
                
            except (ccxt.NetworkError, ccxt.ExchangeNotAvailable) as e:
                retries += 1
                last_error = e
                wait_time = retries * 2  # Exponential backoff
                
                if retry_on_error and retries <= self.max_retries:
                    logger.warning(f"Network error: {str(e)}. Retrying in {wait_time}s... (Attempt {retries}/{self.max_retries})")
                    time.sleep(wait_time)
                else:
                    raise ConnectionError(f"Failed to fetch data after {retries} attempts: {str(e)}")
                    
            except ccxt.ExchangeError as e:
                logger.error(f"Exchange error: {str(e)}")
                raise ValueError(f"Failed to fetch data: Exchange error - {str(e)}")
                
            except Exception as e:
                logger.error(f"Unexpected error: {str(e)}")
                raise Exception(f"Failed to fetch data: {str(e)}")
        
        # If we got here, we've exhausted retries
        raise ConnectionError(f"Failed to fetch data after {self.max_retries} attempts: {str(last_error)}")
    
    def calculate_indicators(self, df: pd.DataFrame, indicators: List[str]) -> pd.DataFrame:
        """
        Calculate technical indicators based on price data.
        
        Args:
            df (pd.DataFrame): OHLCV DataFrame
            indicators (List[str]): List of indicators to calculate
            
        Returns:
            pd.DataFrame: DataFrame with added indicator columns
        """
        analysis = pd.DataFrame()
        analysis['close'] = df['close']
        analysis['open'] = df['open']
        analysis['high'] = df['high']
        analysis['low'] = df['low']
        analysis['volume'] = df['volume']
        
        indicator_map = {
            "RSI": lambda: self._add_rsi(analysis, window=14),
            "MACD": lambda: self._add_macd(analysis),
            "SMA": lambda: self._add_sma(analysis, window=20),
            "EMA": lambda: self._add_ema(analysis, window=20),
            "BB": lambda: self._add_bollinger_bands(analysis, window=20, std=2),
            "ATR": lambda: self._add_atr(analysis, df, window=14),
            "OBV": lambda: self._add_obv(analysis, df),
            "VWAP": lambda: self._add_vwap(analysis, df),
            "Ichimoku": lambda: self._add_ichimoku(analysis),
            "Stochastic": lambda: self._add_stochastic(analysis),
            "Patterns": lambda: self._add_candlestick_patterns(analysis)
        }
        
        # Calculate requested indicators
        for indicator in indicators:
            if indicator in indicator_map:
                try:
                    indicator_map[indicator]()
                except Exception as e:
                    logger.warning(f"Failed to calculate {indicator}: {str(e)}")
            else:
                logger.warning(f"Indicator '{indicator}' not supported.")
        
        return analysis
    
    def _add_rsi(self, df: pd.DataFrame, window: int = 14) -> None:
        """Add Relative Strength Index to DataFrame."""
        df['RSI'] = ta.momentum.RSIIndicator(df['close'], window=window).rsi()
    
    def _add_macd(self, df: pd.DataFrame, fast: int = 12, slow: int = 26, signal: int = 9) -> None:
        """Add MACD indicator to DataFrame."""
        macd_indicator = ta.trend.MACD(
            df['close'], 
            window_fast=fast, 
            window_slow=slow, 
            window_sign=signal
        )
        df['MACD'] = macd_indicator.macd()
        df['MACD_signal'] = macd_indicator.macd_signal()
        df['MACD_histogram'] = macd_indicator.macd_diff()
    
    def _add_sma(self, df: pd.DataFrame, window: int = 20) -> None:
        """Add Simple Moving Average to DataFrame."""
        df[f'SMA_{window}'] = ta.trend.SMAIndicator(df['close'], window=window).sma_indicator()
    
    def _add_ema(self, df: pd.DataFrame, window: int = 20) -> None:
        """Add Exponential Moving Average to DataFrame."""
        df[f'EMA_{window}'] = ta.trend.EMAIndicator(df['close'], window=window).ema_indicator()
    
    def _add_bollinger_bands(self, df: pd.DataFrame, window: int = 20, std: float = 2) -> None:
        """Add Bollinger Bands to DataFrame."""
        bollinger = ta.volatility.BollingerBands(df['close'], window=window, window_dev=std)
        df['BB_upper'] = bollinger.bollinger_hband()
        df['BB_middle'] = bollinger.bollinger_mavg()
        df['BB_lower'] = bollinger.bollinger_lband()
    
    def _add_atr(self, df: pd.DataFrame, ohlc: pd.DataFrame, window: int = 14) -> None:
        """Add Average True Range to DataFrame."""
        df['ATR'] = ta.volatility.AverageTrueRange(
            high=ohlc['high'], 
            low=ohlc['low'], 
            close=ohlc['close'], 
            window=window
        ).average_true_range()
    
    def _add_obv(self, df: pd.DataFrame, ohlc: pd.DataFrame) -> None:
        """Add On-Balance Volume to DataFrame."""
        df['OBV'] = ta.volume.OnBalanceVolumeIndicator(
            close=ohlc['close'], 
            volume=ohlc['volume']
        ).on_balance_volume()
    
    def _add_vwap(self, df: pd.DataFrame, ohlc: pd.DataFrame) -> None:
        """Add Volume Weighted Average Price to DataFrame."""
        try:
            # Reset index to access timestamp for VWAP calculation
            temp_df = ohlc.reset_index()
            # Group by date for daily VWAP
            temp_df['date'] = temp_df['timestamp'].dt.date
            
            typical_price = (temp_df['high'] + temp_df['low'] + temp_df['close']) / 3
            temp_df['VWAP'] = (typical_price * temp_df['volume']).cumsum() / temp_df['volume'].cumsum()
            
            # Set back to original index
            df['VWAP'] = temp_df.set_index('timestamp')['VWAP']
        except Exception as e:
            logger.error(f"VWAP calculation error: {str(e)}")
    
    def _add_ichimoku(self, df: pd.DataFrame) -> None:
        """Add Ichimoku Cloud indicator to DataFrame."""
        try:
            # Tenkan-sen (Conversion Line): (9-period high + 9-period low)/2
            period9_high = df['high'].rolling(window=9).max()
            period9_low = df['low'].rolling(window=9).min()
            df['tenkan_sen'] = (period9_high + period9_low) / 2

            # Kijun-sen (Base Line): (26-period high + 26-period low)/2
            period26_high = df['high'].rolling(window=26).max()
            period26_low = df['low'].rolling(window=26).min()
            df['kijun_sen'] = (period26_high + period26_low) / 2

            # Senkou Span A (Leading Span A): (Conversion Line + Base Line)/2
            df['senkou_span_a'] = ((df['tenkan_sen'] + df['kijun_sen']) / 2).shift(26)

            # Senkou Span B (Leading Span B): (52-period high + 52-period low)/2
            period52_high = df['high'].rolling(window=52).max()
            period52_low = df['low'].rolling(window=52).min()
            df['senkou_span_b'] = ((period52_high + period52_low) / 2).shift(26)

            # Chikou Span (Lagging Span): Close price shifted back 26 periods
            df['chikou_span'] = df['close'].shift(-26)
        except Exception as e:
            logger.error(f"Ichimoku calculation error: {str(e)}")
    
    def _add_stochastic(self, df: pd.DataFrame, k_window: int = 14, d_window: int = 3) -> None:
        """Add Stochastic Oscillator to DataFrame."""
        try:
            stoch = ta.momentum.StochasticOscillator(
                high=df['high'],
                low=df['low'],
                close=df['close'],
                window=k_window,
                smooth_window=d_window
            )
            df['stoch_k'] = stoch.stoch()
            df['stoch_d'] = stoch.stoch_signal()
        except Exception as e:
            logger.error(f"Stochastic calculation error: {str(e)}")
    
    def _add_candlestick_patterns(self, df: pd.DataFrame) -> None:
        """Add candlestick pattern recognition to DataFrame."""
        try:
            # Detect common candlestick patterns
            # Bullish patterns
            df['doji'] = ta.candlestick.doji(df['open'], df['high'], df['low'], df['close'])
            df['hammer'] = ta.candlestick.hammer(df['open'], df['high'], df['low'], df['close'])
            df['morning_star'] = ta.candlestick.morning_star(df['open'], df['high'], df['low'], df['close'])
            
            # Bearish patterns
            df['shooting_star'] = ta.candlestick.shooting_star(df['open'], df['high'], df['low'], df['close'])
            df['evening_star'] = ta.candlestick.evening_star(df['open'], df['high'], df['low'], df['close'])
            df['bearish_harami'] = ta.candlestick.bearish_harami(df['open'], df['high'], df['low'], df['close'])
            
            # Consolidate patterns into single column for easy identification
            df['bullish_pattern'] = (df['doji'] | df['hammer'] | df['morning_star'])
            df['bearish_pattern'] = (df['shooting_star'] | df['evening_star'] | df['bearish_harami'])
        except Exception as e:
            logger.error(f"Pattern recognition error: {str(e)}")
    
    def identify_support_resistance(self, df: pd.DataFrame, window: int = 20, threshold: float = 0.03) -> Tuple[List[float], List[float]]:
        """
        Identify support and resistance levels using pivot points
        
        Args:
            df (pd.DataFrame): OHLCV data
            window (int): Lookback window for pivot identification
            threshold (float): Minimum price change to consider a pivot
            
        Returns:
            Tuple[List[float], List[float]]: Support and resistance levels
        """
        try:
            # Identify pivot highs (resistance)
            pivot_highs = []
            for i in range(window, len(df) - window):
                if all(df['high'].iloc[i] > df['high'].iloc[i-j] for j in range(1, window+1)) and \
                   all(df['high'].iloc[i] > df['high'].iloc[i+j] for j in range(1, window+1)):
                    pivot_highs.append(df['high'].iloc[i])
            
            # Identify pivot lows (support)
            pivot_lows = []
            for i in range(window, len(df) - window):
                if all(df['low'].iloc[i] < df['low'].iloc[i-j] for j in range(1, window+1)) and \
                   all(df['low'].iloc[i] < df['low'].iloc[i+j] for j in range(1, window+1)):
                    pivot_lows.append(df['low'].iloc[i])
            
            # Group close levels
            def group_levels(levels, threshold):
                if not levels:
                    return []
                
                levels = sorted(levels)
                grouped = []
                current_group = [levels[0]]
                
                for level in levels[1:]:
                    if (level - current_group[0]) / current_group[0] <= threshold:
                        current_group.append(level)
                    else:
                        grouped.append(sum(current_group) / len(current_group))
                        current_group = [level]
                
                if current_group:
                    grouped.append(sum(current_group) / len(current_group))
                
                return grouped
            
            return group_levels(pivot_lows, threshold), group_levels(pivot_highs, threshold)
        
        except Exception as e:
            logger.error(f"Support/resistance calculation error: {str(e)}")
            return [], []
    
    def generate_analysis_summary(self, analysis: pd.DataFrame, indicators: List[str]) -> Dict[str, Any]:
        """
        Generate a summary of the technical analysis.
        
        Args:
            analysis (pd.DataFrame): DataFrame with indicator data
            indicators (List[str]): List of indicators used
            
        Returns:
            Dict[str, Any]: Dictionary with analysis results
        """
        summary = {}
        
        try:
            latest = analysis.iloc[-1]
            
            # Overall trend determination
            trend = "Neutral"
            trend_strength = 0
            
            # Using moving averages for trend
            if "SMA" in indicators and "EMA" in indicators:
                sma_cols = [col for col in analysis.columns if 'SMA' in col]
                ema_cols = [col for col in analysis.columns if 'EMA' in col]
                if sma_cols and ema_cols:
                    sma_col = sma_cols[0]
                    ema_col = ema_cols[0]
                    
                    if latest['close'] > latest[sma_col] and latest['close'] > latest[ema_col]:
                        trend = "Bullish"
                        trend_strength += 1
                    elif latest['close'] < latest[sma_col] and latest['close'] < latest[ema_col]:
                        trend = "Bearish"
                        trend_strength += 1
            
            # Using RSI for trend confirmation
            if "RSI" in indicators and not pd.isna(latest.get('RSI', np.nan)):
                rsi_value = latest['RSI']
                if rsi_value > 60:
                    if trend == "Bullish":
                        trend_strength += 1
                    else:
                        trend = "Bullish"
                elif rsi_value < 40:
                    if trend == "Bearish":
                        trend_strength += 1
                    else:
                        trend = "Bearish"
            
            # Using MACD for trend confirmation
            if "MACD" in indicators and not pd.isna(latest.get('MACD', np.nan)):
                if latest['MACD'] > latest['MACD_signal']:
                    if trend == "Bullish":
                        trend_strength += 1
                    else:
                        trend = "Bullish"
                elif latest['MACD'] < latest['MACD_signal']:
                    if trend == "Bearish":
                        trend_strength += 1
                    else:
                        trend = "Bearish"
            
            # Store trend information
            summary['trend'] = trend
            summary['trend_strength'] = f"{trend_strength}/3" if trend_strength > 0 else "Weak"
            
            # RSI analysis
            if "RSI" in indicators and not pd.isna(latest.get('RSI', np.nan)):
                rsi_value = latest['RSI']
                
                # Check RSI divergence
                has_divergence = False
                divergence_type = None
                
                if len(analysis) > 20:  # Need enough data for divergence
                    # Find recent price high/low
                    price_section = analysis['close'].iloc[-20:]
                    rsi_section = analysis['RSI'].iloc[-20:]
                    
                    price_high_idx = price_section.idxmax()
                    price_low_idx = price_section.idxmin()
                    rsi_high_idx = rsi_section.idxmax()
                    rsi_low_idx = rsi_section.idxmin()
                    
                    # Bearish divergence: price makes higher high, RSI makes lower high
                    if price_high_idx != rsi_high_idx and price_section.max() > price_section.iloc[0]:
                        has_divergence = True
                        divergence_type = "Bearish"
                    
                    # Bullish divergence: price makes lower low, RSI makes higher low
                    if price_low_idx != rsi_low_idx and price_section.min() < price_section.iloc[0]:
                        has_divergence = True
                        divergence_type = "Bullish"
                
                summary['RSI'] = {
                    'value': round(rsi_value, 2),
                    'interpretation': "Overbought" if rsi_value > 70 else "Oversold" if rsi_value < 30 else "Neutral",
                    'has_divergence': has_divergence,
                    'divergence_type': divergence_type
                }
            
            # MACD analysis
            if "MACD" in indicators and not pd.isna(latest.get('MACD', np.nan)):
                macd_value = latest['MACD']
                signal_value = latest['MACD_signal']
                cross_direction = None
                
                # Check for recent crossover (past 5 periods)
                for i in range(min(5, len(analysis) - 1)):
                    prev_idx = -2 - i
                    if (analysis['MACD'].iloc[prev_idx] <= analysis['MACD_signal'].iloc[prev_idx] and 
                        macd_value > signal_value):
                        cross_direction = "Bullish crossover"
                        break
                    elif (analysis['MACD'].iloc[prev_idx] >= analysis['MACD_signal'].iloc[prev_idx] and 
                          macd_value < signal_value):
                        cross_direction = "Bearish crossover"
                        break
                
                # Check for histogram momentum
                histogram_momentum = "Increasing" if latest['MACD_histogram'] > analysis['MACD_histogram'].iloc[-2] else "Decreasing"
                
                summary['MACD'] = {
                    'value': round(macd_value, 2),
                    'signal': round(signal_value, 2),
                    'histogram': round(latest['MACD_histogram'], 2),
                    'histogram_momentum': histogram_momentum,
                    'interpretation': cross_direction if cross_direction else "Neutral"
                }
            
            # Bollinger Bands analysis
            if "BB" in indicators and "BB_upper" in analysis.columns:
                # Calculate bandwidth
                bandwidth = (latest['BB_upper'] - latest['BB_lower']) / latest['BB_middle'] * 100
                if latest['close'] > latest['BB_upper']:
                    bb_position = "Above upper band (potentially overbought)"
                elif latest['close'] < latest['BB_lower']:
                    bb_position = "Below lower band (potentially oversold)"
                else:
                    # Calculate position within bands as percentage
                    band_width = latest['BB_upper'] - latest['BB_lower']
                    if band_width > 0:
                        position = (latest['close'] - latest['BB_lower']) / band_width * 100
                        bb_position = f"Within bands ({round(position, 1)}% from lower band)"
                    else:
                        bb_position = "Within bands"
                
                # Check for BB squeeze (narrowing bands)
                is_squeeze = False
                if len(analysis) > 20:
                    prev_bandwidth = (analysis['BB_upper'].iloc[-20] - analysis['BB_lower'].iloc[-20]) / analysis['BB_middle'].iloc[-20] * 100
                    is_squeeze = bandwidth < prev_bandwidth * 0.8  # 20% narrower bands
                
                summary['Bollinger_Bands'] = {
                    'upper': round(latest['BB_upper'], 2),
                    'middle': round(latest['BB_middle'], 2),
                    'lower': round(latest['BB_lower'], 2),
                    'bandwidth': round(bandwidth, 2),
                    'position': bb_position,
                    'squeeze': is_squeeze
                }
            
            # Support and Resistance levels
            support, resistance = self.identify_support_resistance(analysis)
            summary['Support'] = support
            summary['Resistance'] = resistance
            
            return summary
        
        except Exception as e:
            logger.error(f"Error generating summary: {str(e)}")
            return summary
    
    def plot_chart(self, df: pd.DataFrame, ticker: str, timeframe: str, analysis: pd.DataFrame, indicators: List[str]) -> str:
        """
        Plot candlestick chart with indicators and save chart image.
        
        Returns:
            str: File path of saved chart image.
        """
        try:
            plt.figure(figsize=(12,8))
            # Plot close price
            plt.plot(df.index, df['close'], label='Close Price', color='blue')
            
            # Plot SMA and EMA if available
            for col in df.columns:
                if 'SMA' in col or 'EMA' in col:
                    plt.plot(df.index, df[col], label=col)
            
            # Format x-axis with date labels
            plt.gca().xaxis.set_major_formatter(mdates.DateFormatter('%Y-%m-%d'))
            plt.gca().xaxis.set_major_locator(mdates.AutoDateLocator())
            plt.gcf().autofmt_xdate()
            
            plt.title(f"{ticker} Price Chart ({timeframe})")
            plt.xlabel("Date")
            plt.ylabel("Price")
            plt.legend()
            plt.grid(True)
            
            # Save chart
            timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
            chart_filename = f"{ticker.replace('/', '_')}_{timeframe}_{timestamp}.png"
            chart_path = os.path.join(self.output_dir, chart_filename)
            plt.savefig(chart_path)
            plt.close()
            return chart_path
        except Exception as e:
            logger.error(f"Chart plotting error: {str(e)}")
            return ""
    
    def analyze(self, ticker: str, timeframe: str, indicators: List[str], days: int = 30) -> AnalysisResult:
        """
        Perform full analysis: fetch data, calculate indicators, generate summary and chart.
        
        Returns:
            AnalysisResult: Structured analysis result.
        """
        try:
            df = self.fetch_data(ticker, timeframe, days=days)
            analysis_df = self.calculate_indicators(df, indicators)
            summary = self.generate_analysis_summary(analysis_df, indicators)
            chart_path = self.plot_chart(df, ticker, timeframe, analysis_df, indicators)
            period = f"Last {days} days"
            timestamp = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
            result = AnalysisResult(
                ticker=ticker,
                timeframe=timeframe,
                summary=summary,
                chart_path=chart_path,
                indicators_used=indicators,
                data_points=len(df),
                period=period,
                timestamp=timestamp
            )
            return result
        except Exception as e:
            logger.error(f"Analysis failed: {traceback.format_exc()}")
            raise e

if __name__ == "__main__":
    analyzer = CryptoAnalyzer(exchange_name="binance", output_dir="./charts", rate_limit_pause=1.0, max_retries=3)
    ticker = "BTC/USDT"
    timeframe = "1d"
    indicators = ["RSI", "MACD", "SMA", "EMA", "BB", "ATR", "OBV", "VWAP", "Ichimoku", "Stochastic", "Patterns"]
    try:
        result = analyzer.analyze(ticker, timeframe, indicators, days=90)
        print(result.get_summary_text())
        print(result.to_json())
    except Exception as e:
        print(f"Error during analysis: {e}")