quantalogic 0.33.4__py3-none-any.whl → 0.40.0__py3-none-any.whl

quantalogic/tools/finance/ccxt_tool.py

@@ -0,0 +1,373 @@
+import json
+from concurrent.futures import ThreadPoolExecutor
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Any, ClassVar, Dict, List
+
+import ccxt.async_support as ccxt
+import pandas as pd
+import ta
+from loguru import logger
+
+from quantalogic.tools import Tool, ToolArgument
+
+
+@dataclass
+class MarketData:
+    """Container for market data and analysis."""
+    symbol: str
+    exchange: str
+    timeframe: str
+    data: pd.DataFrame
+    indicators: Dict[str, pd.Series] = None
+    patterns: Dict[str, pd.Series] = None
+    levels: Dict[str, float] = None
+    volume_profile: Dict[str, float] = None
+
+class CCXTTool(Tool):
+    """Advanced cryptocurrency trading and analysis tool using CCXT."""
+
+    name: ClassVar[str] = "ccxt_tool"
+    description: ClassVar[str] = "Enhanced cryptocurrency trading and analysis tool using CCXT"
+    
+    TIMEFRAMES: ClassVar[List[str]] = [
+        '1m', '3m', '5m', '15m', '30m', '1h', '2h', '4h', '6h', '8h', '12h', '1d', '3d', '1w', '1M'
+    ]
+
+    arguments: ClassVar[list[ToolArgument]] = [
+        ToolArgument(
+            name="exchange_ids",
+            arg_type="string",
+            description="Comma-separated exchange IDs (e.g., 'binance,kucoin')",
+            required=True
+        ),
+        ToolArgument(
+            name="symbols",
+            arg_type="string",
+            description="Comma-separated trading pairs (e.g., 'BTC/USDT,ETH/USDT')",
+            required=True
+        ),
+        ToolArgument(
+            name="timeframe",
+            arg_type="string",
+            description="Time interval for data",
+            required=False,
+            default="1h"
+        ),
+        ToolArgument(
+            name="lookback_periods",
+            arg_type="string",
+            description="Number of periods to analyze",
+            required=False,
+            default="500"
+        ),
+        ToolArgument(
+            name="analysis_types",
+            arg_type="string",
+            description="Comma-separated analysis types (technical,patterns,volume,all)",
+            required=False,
+            default="all"
+        ),
+        ToolArgument(
+            name="credentials_path",
+            arg_type="string",
+            description="Path to exchange credentials file",
+            required=False,
+            default="config/exchange_credentials.json"
+        )
+    ]
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+        self.exchanges: Dict[str, ccxt.Exchange] = {}
+        self.cache = {}
+        self.executor = ThreadPoolExecutor(max_workers=4)
+
+    def validate_arguments(self, **kwargs) -> bool:
+        """Validate tool arguments."""
+        try:
+            # Validate required arguments
+            required_args = [arg.name for arg in self.arguments if arg.required]
+            for arg in required_args:
+                if arg not in kwargs:
+                    raise ValueError(f"Missing required argument: {arg}")
+
+            # Validate timeframe
+            if 'timeframe' in kwargs and kwargs['timeframe'] not in self.TIMEFRAMES:
+                raise ValueError(f"Invalid timeframe: {kwargs['timeframe']}")
+
+            # Validate exchange IDs format
+            if 'exchange_ids' in kwargs:
+                exchange_ids = kwargs['exchange_ids'].split(',')
+                if not all(exchange_id.strip() for exchange_id in exchange_ids):
+                    raise ValueError("Invalid exchange IDs format")
+
+            # Validate trading pairs format
+            if 'symbols' in kwargs:
+                symbols = kwargs['symbols'].split(',')
+                if not all('/' in symbol for symbol in symbols):
+                    raise ValueError("Invalid trading pair format. Must be in format BASE/QUOTE")
+
+            return True
+        except Exception as e:
+            logger.error(f"Error validating arguments: {e}")
+            raise
+
+    async def _initialize_exchanges(self, exchange_ids: List[str], credentials_path: str) -> None:
+        """Initialize exchange instances with credentials if available."""
+        try:
+            creds = {}
+            creds_path = Path(credentials_path)
+            if creds_path.exists():
+                with open(creds_path) as f:
+                    creds = json.load(f)
+
+            for exchange_id in exchange_ids:
+                exchange_class = getattr(ccxt, exchange_id)
+                exchange_creds = creds.get(exchange_id, {})
+                
+                self.exchanges[exchange_id] = exchange_class({
+                    'apiKey': exchange_creds.get('api_key'),
+                    'secret': exchange_creds.get('secret'),
+                    'password': exchange_creds.get('password'),
+                    'enableRateLimit': True,
+                    'options': {'defaultType': 'spot'}
+                })
+
+                # Load markets for symbol validation
+                await self.exchanges[exchange_id].load_markets()
+                
+        except Exception as e:
+            logger.error(f"Error initializing exchanges: {e}")
+            raise
+
+    async def _fetch_ohlcv(
+        self,
+        exchange_id: str,
+        symbol: str,
+        timeframe: str,
+        limit: int
+    ) -> MarketData:
+        """Fetch OHLCV data from exchange."""
+        try:
+            exchange = self.exchanges[exchange_id]
+            
+            # Fetch OHLCV data
+            ohlcv = await exchange.fetch_ohlcv(symbol, timeframe, limit=limit)
+            
+            # 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 MarketData(
+                symbol=symbol,
+                exchange=exchange_id,
+                timeframe=timeframe,
+                data=df
+            )
+            
+        except Exception as e:
+            logger.error(f"Error fetching OHLCV data for {symbol} on {exchange_id}: {e}")
+            raise
+
+    async def _fetch_order_book(self, exchange_id: str, symbol: str, limit: int = 100) -> Dict:
+        """Fetch order book data."""
+        try:
+            exchange = self.exchanges[exchange_id]
+            order_book = await exchange.fetch_order_book(symbol, limit)
+            
+            return {
+                'bids': order_book['bids'],
+                'asks': order_book['asks'],
+                'timestamp': order_book['timestamp'],
+                'datetime': order_book['datetime'],
+                'nonce': order_book.get('nonce')
+            }
+        except Exception as e:
+            logger.error(f"Error fetching order book for {symbol} on {exchange_id}: {e}")
+            raise
+
+    def _calculate_technical_indicators(self, market_data: MarketData) -> None:
+        """Calculate comprehensive technical indicators."""
+        df = market_data.data
+        indicators = {}
+        
+        try:
+            # Trend Indicators
+            indicators['sma_20'] = ta.trend.sma_indicator(df['close'], 20)
+            indicators['sma_50'] = ta.trend.sma_indicator(df['close'], 50)
+            indicators['sma_200'] = ta.trend.sma_indicator(df['close'], 200)
+            indicators['ema_12'] = ta.trend.ema_indicator(df['close'], 12)
+            indicators['ema_26'] = ta.trend.ema_indicator(df['close'], 26)
+            indicators['macd'] = ta.trend.macd(df['close'])
+            indicators['macd_signal'] = ta.trend.macd_signal(df['close'])
+            indicators['macd_diff'] = ta.trend.macd_diff(df['close'])
+            indicators['adx'] = ta.trend.adx(df['high'], df['low'], df['close'])
+            
+            # Momentum Indicators
+            indicators['rsi'] = ta.momentum.rsi(df['close'])
+            indicators['stoch'] = ta.momentum.stoch(df['high'], df['low'], df['close'])
+            indicators['stoch_signal'] = ta.momentum.stoch_signal(df['high'], df['low'], df['close'])
+            indicators['williams_r'] = ta.momentum.williams_r(df['high'], df['low'], df['close'])
+            
+            # Volatility Indicators
+            indicators['bbands_upper'] = ta.volatility.bollinger_hband(df['close'])
+            indicators['bbands_lower'] = ta.volatility.bollinger_lband(df['close'])
+            indicators['bbands_middle'] = ta.volatility.bollinger_mavg(df['close'])
+            indicators['atr'] = ta.volatility.average_true_range(df['high'], df['low'], df['close'])
+            
+            # Volume Indicators
+            indicators['obv'] = ta.volume.on_balance_volume(df['close'], df['volume'])
+            indicators['mfi'] = ta.volume.money_flow_index(df['high'], df['low'], df['close'], df['volume'])
+            indicators['vwap'] = self._calculate_vwap(df)
+            
+            market_data.indicators = indicators
+            
+        except Exception as e:
+            logger.error(f"Error calculating technical indicators: {e}")
+            raise
+
+    def _calculate_vwap(self, df: pd.DataFrame) -> pd.Series:
+        """Calculate Volume Weighted Average Price."""
+        v = df['volume']
+        tp = (df['high'] + df['low'] + df['close']) / 3
+        return (tp * v).cumsum() / v.cumsum()
+
+    def _identify_patterns(self, market_data: MarketData) -> None:
+        """Identify chart patterns and candlestick patterns."""
+        df = market_data.data
+        patterns = {}
+        
+        try:
+            # Candlestick Patterns
+            patterns['doji'] = ta.candlestick.doji(df['open'], df['high'], df['low'], df['close'])
+            patterns['hammer'] = ta.candlestick.hammer(df['open'], df['high'], df['low'], df['close'])
+            patterns['shooting_star'] = ta.candlestick.shooting_star(df['open'], df['high'], df['low'], df['close'])
+            patterns['morning_star'] = ta.candlestick.morning_star(df['open'], df['high'], df['low'], df['close'])
+            patterns['evening_star'] = ta.candlestick.evening_star(df['open'], df['high'], df['low'], df['close'])
+            
+            # Custom Pattern Detection
+            patterns['double_top'] = self._detect_double_top(df)
+            patterns['double_bottom'] = self._detect_double_bottom(df)
+            patterns['head_shoulders'] = self._detect_head_shoulders(df)
+            
+            market_data.patterns = patterns
+            
+        except Exception as e:
+            logger.error(f"Error identifying patterns: {e}")
+            raise
+
+    def _detect_double_top(self, df: pd.DataFrame) -> pd.Series:
+        """Detect double top pattern."""
+        window = 20
+        peaks = df['high'].rolling(window, center=True).apply(
+            lambda x: 1 if x.iloc[window//2] == max(x) else 0
+        )
+        return peaks
+
+    def _detect_double_bottom(self, df: pd.DataFrame) -> pd.Series:
+        """Detect double bottom pattern."""
+        window = 20
+        troughs = df['low'].rolling(window, center=True).apply(
+            lambda x: 1 if x.iloc[window//2] == min(x) else 0
+        )
+        return troughs
+
+    def _detect_head_shoulders(self, df: pd.DataFrame) -> pd.Series:
+        """Detect head and shoulders pattern."""
+        window = 30
+        result = pd.Series(0, index=df.index)
+        
+        for i in range(window, len(df)-window):
+            left = df['high'].iloc[i-window:i].max()
+            head = df['high'].iloc[i]
+            right = df['high'].iloc[i:i+window].max()
+            
+            if head > left and head > right and abs(left - right) < 0.1 * head:
+                result.iloc[i] = 1
+                
+        return result
+
+    def _analyze_volume_profile(self, market_data: MarketData) -> None:
+        """Analyze volume profile and identify key price levels."""
+        df = market_data.data
+        
+        try:
+            # Calculate volume profile
+            price_bins = pd.qcut(df['close'], q=50, duplicates='drop')
+            volume_profile = df.groupby(price_bins)['volume'].sum()
+            
+            # Identify high volume nodes
+            high_volume_levels = volume_profile.nlargest(5)
+            
+            market_data.volume_profile = {
+                'volume_by_price': volume_profile.to_dict(),
+                'high_volume_levels': high_volume_levels.to_dict()
+            }
+            
+        except Exception as e:
+            logger.error(f"Error analyzing volume profile: {e}")
+            raise
+
+    async def execute(self, **kwargs) -> Dict[str, Any]:
+        """Execute the CCXT tool with comprehensive analysis."""
+        try:
+            self.validate_arguments(**kwargs)
+            
+            exchange_ids = kwargs['exchange_ids'].split(',')  # Split comma-separated string
+            symbols = kwargs['symbols'].split(',')  # Split comma-separated string
+            timeframe = kwargs.get('timeframe', '1h')
+            lookback_periods = int(kwargs.get('lookback_periods', '500'))  # Convert to int
+            analysis_types = kwargs.get('analysis_types', 'all').split(',')  # Split comma-separated string
+            credentials_path = kwargs.get('credentials_path', 'config/exchange_credentials.json')
+            
+            # Initialize exchanges
+            await self._initialize_exchanges(exchange_ids, credentials_path)
+            
+            # Fetch data and analyze for each symbol on each exchange
+            results = {}
+            for exchange_id in exchange_ids:
+                for symbol in symbols:
+                    # Fetch OHLCV data
+                    market_data = await self._fetch_ohlcv(
+                        exchange_id, symbol, timeframe, lookback_periods
+                    )
+                    
+                    # Perform requested analyses
+                    if 'all' in analysis_types or 'technical' in analysis_types:
+                        self._calculate_technical_indicators(market_data)
+                    
+                    if 'all' in analysis_types or 'patterns' in analysis_types:
+                        self._identify_patterns(market_data)
+                    
+                    if 'all' in analysis_types or 'volume' in analysis_types:
+                        self._analyze_volume_profile(market_data)
+                    
+                    if 'all' in analysis_types or 'orderbook' in analysis_types:
+                        order_book = await self._fetch_order_book(exchange_id, symbol)
+                    else:
+                        order_book = None
+                    
+                    # Format results
+                    key = f"{exchange_id}:{symbol}"
+                    results[key] = {
+                        'market_data': market_data.data.to_dict(orient='records'),
+                        'indicators': {k: v.to_dict() for k, v in market_data.indicators.items()} if market_data.indicators else None,
+                        'patterns': {k: v.to_dict() for k, v in market_data.patterns.items()} if market_data.patterns else None,
+                        'volume_profile': market_data.volume_profile,
+                        'order_book': order_book
+                    }
+            
+            # Close exchange connections
+            for exchange in self.exchanges.values():
+                await exchange.close()
+            
+            return results
+            
+        except Exception as e:
+            logger.error(f"Error executing CCXT tool: {e}")
+            raise