sigma-terminal 2.0.2__py3-none-any.whl → 3.3.0__py3-none-any.whl

sigma/monitoring.py

@@ -0,0 +1,666 @@
+"""Monitoring system - Watchlists, alerts, drift detection."""
+
+import asyncio
+import json
+import os
+from datetime import datetime, date, timedelta
+from pathlib import Path
+from typing import Any, Callable, Dict, List, Optional, Union
+from enum import Enum
+
+import numpy as np
+import pandas as pd
+from pydantic import BaseModel, Field
+
+
+# ============================================================================
+# DATA MODELS
+# ============================================================================
+
+class AlertType(str, Enum):
+    """Types of alerts."""
+    PRICE_ABOVE = "price_above"
+    PRICE_BELOW = "price_below"
+    PERCENT_CHANGE = "percent_change"
+    VOLUME_SPIKE = "volume_spike"
+    VOLATILITY_SPIKE = "volatility_spike"
+    MOVING_AVERAGE_CROSS = "ma_cross"
+    RSI_OVERSOLD = "rsi_oversold"
+    RSI_OVERBOUGHT = "rsi_overbought"
+    DRAWDOWN = "drawdown"
+    NEW_HIGH = "new_high"
+    NEW_LOW = "new_low"
+    CORRELATION_BREAK = "correlation_break"
+    DRIFT_DETECTED = "drift_detected"
+
+
+class AlertPriority(str, Enum):
+    """Alert priority levels."""
+    LOW = "low"
+    MEDIUM = "medium"
+    HIGH = "high"
+    CRITICAL = "critical"
+
+
+class Alert(BaseModel):
+    """Alert configuration."""
+    
+    id: str = Field(default_factory=lambda: datetime.now().strftime("%Y%m%d%H%M%S%f"))
+    symbol: str
+    alert_type: AlertType
+    threshold: float
+    priority: AlertPriority = AlertPriority.MEDIUM
+    enabled: bool = True
+    created_at: datetime = Field(default_factory=datetime.now)
+    last_triggered: Optional[datetime] = None
+    trigger_count: int = 0
+    cooldown_minutes: int = 60  # Don't re-trigger within this period
+    message: Optional[str] = None
+    metadata: Dict[str, Any] = Field(default_factory=dict)
+
+
+class AlertNotification(BaseModel):
+    """Alert notification event."""
+    
+    alert_id: str
+    symbol: str
+    alert_type: AlertType
+    priority: AlertPriority
+    message: str
+    current_value: float
+    threshold: float
+    triggered_at: datetime = Field(default_factory=datetime.now)
+
+
+class WatchlistItem(BaseModel):
+    """Item in a watchlist."""
+    
+    symbol: str
+    added_at: datetime = Field(default_factory=datetime.now)
+    notes: Optional[str] = None
+    target_price: Optional[float] = None
+    stop_price: Optional[float] = None
+    tags: List[str] = Field(default_factory=list)
+    alerts: List[str] = Field(default_factory=list)  # Alert IDs
+
+
+class Watchlist(BaseModel):
+    """A watchlist of securities."""
+    
+    id: str = Field(default_factory=lambda: datetime.now().strftime("%Y%m%d%H%M%S"))
+    name: str
+    description: Optional[str] = None
+    items: List[WatchlistItem] = Field(default_factory=list)
+    created_at: datetime = Field(default_factory=datetime.now)
+    updated_at: datetime = Field(default_factory=datetime.now)
+
+
+# ============================================================================
+# ALERT ENGINE
+# ============================================================================
+
+class AlertEngine:
+    """
+    Monitor markets and trigger alerts.
+    Supports price alerts, technical alerts, and custom conditions.
+    """
+    
+    def __init__(self, storage_dir: str = None):
+        self.storage_dir = Path(storage_dir or os.path.expanduser("~/.sigma/alerts"))
+        self.storage_dir.mkdir(parents=True, exist_ok=True)
+        self.alerts: Dict[str, Alert] = {}
+        self.notification_handlers: List[Callable[[AlertNotification], None]] = []
+        self._load_alerts()
+    
+    def _load_alerts(self):
+        """Load alerts from storage."""
+        
+        alerts_file = self.storage_dir / "alerts.json"
+        if alerts_file.exists():
+            data = json.loads(alerts_file.read_text())
+            for alert_data in data:
+                alert = Alert(**alert_data)
+                self.alerts[alert.id] = alert
+    
+    def _save_alerts(self):
+        """Save alerts to storage."""
+        
+        alerts_file = self.storage_dir / "alerts.json"
+        data = [alert.model_dump() for alert in self.alerts.values()]
+        alerts_file.write_text(json.dumps(data, indent=2, default=str))
+    
+    def add_alert(self, alert: Alert) -> str:
+        """Add a new alert."""
+        
+        self.alerts[alert.id] = alert
+        self._save_alerts()
+        return alert.id
+    
+    def remove_alert(self, alert_id: str) -> bool:
+        """Remove an alert."""
+        
+        if alert_id in self.alerts:
+            del self.alerts[alert_id]
+            self._save_alerts()
+            return True
+        return False
+    
+    def enable_alert(self, alert_id: str, enabled: bool = True):
+        """Enable or disable an alert."""
+        
+        if alert_id in self.alerts:
+            self.alerts[alert_id].enabled = enabled
+            self._save_alerts()
+    
+    def add_notification_handler(self, handler: Callable[[AlertNotification], None]):
+        """Add a notification handler."""
+        
+        self.notification_handlers.append(handler)
+    
+    async def check_alerts(self, market_data: Dict[str, Dict[str, Any]]) -> List[AlertNotification]:
+        """
+        Check all alerts against current market data.
+        
+        Args:
+            market_data: Dict of symbol -> current data
+        
+        Returns:
+            List of triggered notifications
+        """
+        
+        notifications = []
+        now = datetime.now()
+        
+        for alert_id, alert in self.alerts.items():
+            if not alert.enabled:
+                continue
+            
+            # Check cooldown
+            if alert.last_triggered:
+                cooldown_end = alert.last_triggered + timedelta(minutes=alert.cooldown_minutes)
+                if now < cooldown_end:
+                    continue
+            
+            # Get market data for symbol
+            data = market_data.get(alert.symbol)
+            if not data:
+                continue
+            
+            # Check alert condition
+            triggered, current_value, message = self._check_alert_condition(alert, data)
+            
+            if triggered:
+                # Create notification
+                notification = AlertNotification(
+                    alert_id=alert.id,
+                    symbol=alert.symbol,
+                    alert_type=alert.alert_type,
+                    priority=alert.priority,
+                    message=message or alert.message or f"{alert.alert_type.value} triggered",
+                    current_value=current_value,
+                    threshold=alert.threshold,
+                )
+                
+                notifications.append(notification)
+                
+                # Update alert
+                alert.last_triggered = now
+                alert.trigger_count += 1
+                
+                # Notify handlers
+                for handler in self.notification_handlers:
+                    try:
+                        handler(notification)
+                    except Exception as e:
+                        print(f"Notification handler error: {e}")
+        
+        self._save_alerts()
+        return notifications
+    
+    def _check_alert_condition(
+        self,
+        alert: Alert,
+        data: Dict[str, Any],
+    ) -> tuple[bool, float, str]:
+        """Check if alert condition is met."""
+        
+        current_price = data.get("price", data.get("close", 0))
+        
+        if alert.alert_type == AlertType.PRICE_ABOVE:
+            if current_price > alert.threshold:
+                return True, current_price, f"{alert.symbol} above ${alert.threshold:.2f} (${current_price:.2f})"
+        
+        elif alert.alert_type == AlertType.PRICE_BELOW:
+            if current_price < alert.threshold:
+                return True, current_price, f"{alert.symbol} below ${alert.threshold:.2f} (${current_price:.2f})"
+        
+        elif alert.alert_type == AlertType.PERCENT_CHANGE:
+            change = data.get("change_percent", 0)
+            if abs(change) > alert.threshold:
+                direction = "up" if change > 0 else "down"
+                return True, change, f"{alert.symbol} {direction} {abs(change):.1f}%"
+        
+        elif alert.alert_type == AlertType.VOLUME_SPIKE:
+            volume = data.get("volume", 0)
+            avg_volume = data.get("avg_volume", volume)
+            if avg_volume > 0 and volume / avg_volume > alert.threshold:
+                multiple = volume / avg_volume
+                return True, multiple, f"{alert.symbol} volume {multiple:.1f}x average"
+        
+        elif alert.alert_type == AlertType.VOLATILITY_SPIKE:
+            volatility = data.get("volatility", 0)
+            if volatility > alert.threshold:
+                return True, volatility, f"{alert.symbol} volatility at {volatility:.1%}"
+        
+        elif alert.alert_type == AlertType.RSI_OVERSOLD:
+            rsi = data.get("rsi", 50)
+            if rsi < alert.threshold:
+                return True, rsi, f"{alert.symbol} RSI oversold at {rsi:.1f}"
+        
+        elif alert.alert_type == AlertType.RSI_OVERBOUGHT:
+            rsi = data.get("rsi", 50)
+            if rsi > alert.threshold:
+                return True, rsi, f"{alert.symbol} RSI overbought at {rsi:.1f}"
+        
+        elif alert.alert_type == AlertType.DRAWDOWN:
+            drawdown = data.get("drawdown", 0)
+            if abs(drawdown) > alert.threshold:
+                return True, drawdown, f"{alert.symbol} drawdown at {abs(drawdown):.1%}"
+        
+        elif alert.alert_type == AlertType.NEW_HIGH:
+            high_52w = data.get("52w_high", float('inf'))
+            if current_price >= high_52w * (1 - alert.threshold / 100):
+                return True, current_price, f"{alert.symbol} near 52-week high"
+        
+        elif alert.alert_type == AlertType.NEW_LOW:
+            low_52w = data.get("52w_low", 0)
+            if current_price <= low_52w * (1 + alert.threshold / 100):
+                return True, current_price, f"{alert.symbol} near 52-week low"
+        
+        return False, 0, ""
+    
+    def get_alerts_for_symbol(self, symbol: str) -> List[Alert]:
+        """Get all alerts for a symbol."""
+        
+        return [a for a in self.alerts.values() if a.symbol == symbol]
+    
+    def get_active_alerts(self) -> List[Alert]:
+        """Get all active alerts."""
+        
+        return [a for a in self.alerts.values() if a.enabled]
+
+
+# ============================================================================
+# WATCHLIST MANAGER
+# ============================================================================
+
+class WatchlistManager:
+    """Manage watchlists."""
+    
+    def __init__(self, storage_dir: str = None):
+        self.storage_dir = Path(storage_dir or os.path.expanduser("~/.sigma/watchlists"))
+        self.storage_dir.mkdir(parents=True, exist_ok=True)
+        self.watchlists: Dict[str, Watchlist] = {}
+        self._load_watchlists()
+    
+    def _load_watchlists(self):
+        """Load watchlists from storage."""
+        
+        for file in self.storage_dir.glob("*.json"):
+            try:
+                data = json.loads(file.read_text())
+                watchlist = Watchlist(**data)
+                self.watchlists[watchlist.id] = watchlist
+            except Exception as e:
+                print(f"Error loading watchlist {file}: {e}")
+    
+    def _save_watchlist(self, watchlist: Watchlist):
+        """Save a watchlist to storage."""
+        
+        filepath = self.storage_dir / f"{watchlist.id}.json"
+        filepath.write_text(watchlist.model_dump_json(indent=2))
+    
+    def create_watchlist(self, name: str, description: str = None) -> Watchlist:
+        """Create a new watchlist."""
+        
+        watchlist = Watchlist(name=name, description=description)
+        self.watchlists[watchlist.id] = watchlist
+        self._save_watchlist(watchlist)
+        return watchlist
+    
+    def delete_watchlist(self, watchlist_id: str) -> bool:
+        """Delete a watchlist."""
+        
+        if watchlist_id in self.watchlists:
+            del self.watchlists[watchlist_id]
+            filepath = self.storage_dir / f"{watchlist_id}.json"
+            if filepath.exists():
+                filepath.unlink()
+            return True
+        return False
+    
+    def add_to_watchlist(
+        self,
+        watchlist_id: str,
+        symbol: str,
+        notes: str = None,
+        target_price: float = None,
+        stop_price: float = None,
+        tags: List[str] = None,
+    ) -> bool:
+        """Add a symbol to a watchlist."""
+        
+        if watchlist_id not in self.watchlists:
+            return False
+        
+        watchlist = self.watchlists[watchlist_id]
+        
+        # Check if already in watchlist
+        if any(item.symbol == symbol for item in watchlist.items):
+            return False
+        
+        item = WatchlistItem(
+            symbol=symbol,
+            notes=notes,
+            target_price=target_price,
+            stop_price=stop_price,
+            tags=tags or [],
+        )
+        
+        watchlist.items.append(item)
+        watchlist.updated_at = datetime.now()
+        self._save_watchlist(watchlist)
+        
+        return True
+    
+    def remove_from_watchlist(self, watchlist_id: str, symbol: str) -> bool:
+        """Remove a symbol from a watchlist."""
+        
+        if watchlist_id not in self.watchlists:
+            return False
+        
+        watchlist = self.watchlists[watchlist_id]
+        original_length = len(watchlist.items)
+        watchlist.items = [item for item in watchlist.items if item.symbol != symbol]
+        
+        if len(watchlist.items) < original_length:
+            watchlist.updated_at = datetime.now()
+            self._save_watchlist(watchlist)
+            return True
+        
+        return False
+    
+    def get_watchlist(self, watchlist_id: str) -> Optional[Watchlist]:
+        """Get a watchlist by ID."""
+        
+        return self.watchlists.get(watchlist_id)
+    
+    def get_all_watchlists(self) -> List[Watchlist]:
+        """Get all watchlists."""
+        
+        return list(self.watchlists.values())
+    
+    def get_all_symbols(self) -> List[str]:
+        """Get all unique symbols across all watchlists."""
+        
+        symbols = set()
+        for watchlist in self.watchlists.values():
+            for item in watchlist.items:
+                symbols.add(item.symbol)
+        return list(symbols)
+
+
+# ============================================================================
+# DRIFT DETECTOR
+# ============================================================================
+
+class DriftDetector:
+    """
+    Detect drift in portfolios and strategies.
+    Monitors for changes in:
+    - Portfolio weights
+    - Risk characteristics
+    - Return patterns
+    - Correlations
+    """
+    
+    def __init__(
+        self,
+        weight_threshold: float = 0.05,
+        volatility_threshold: float = 0.5,
+        correlation_threshold: float = 0.3,
+    ):
+        self.weight_threshold = weight_threshold
+        self.volatility_threshold = volatility_threshold
+        self.correlation_threshold = correlation_threshold
+    
+    def check_weight_drift(
+        self,
+        current_weights: Dict[str, float],
+        target_weights: Dict[str, float],
+    ) -> Dict[str, Any]:
+        """Check for portfolio weight drift."""
+        
+        drifts = {}
+        max_drift = 0
+        
+        all_assets = set(current_weights.keys()) | set(target_weights.keys())
+        
+        for asset in all_assets:
+            current = current_weights.get(asset, 0)
+            target = target_weights.get(asset, 0)
+            drift = current - target
+            drifts[asset] = drift
+            max_drift = max(max_drift, abs(drift))
+        
+        needs_rebalance = max_drift > self.weight_threshold
+        
+        return {
+            "drifts": drifts,
+            "max_drift": max_drift,
+            "needs_rebalance": needs_rebalance,
+            "threshold": self.weight_threshold,
+        }
+    
+    def check_volatility_drift(
+        self,
+        current_vol: float,
+        historical_vol: float,
+    ) -> Dict[str, Any]:
+        """Check for volatility regime change."""
+        
+        if historical_vol == 0:
+            return {"drift_detected": False, "reason": "No historical data"}
+        
+        vol_change = (current_vol - historical_vol) / historical_vol
+        drift_detected = abs(vol_change) > self.volatility_threshold
+        
+        return {
+            "current_vol": current_vol,
+            "historical_vol": historical_vol,
+            "change": vol_change,
+            "drift_detected": drift_detected,
+            "direction": "up" if vol_change > 0 else "down",
+            "threshold": self.volatility_threshold,
+        }
+    
+    def check_correlation_drift(
+        self,
+        current_corr: pd.DataFrame,
+        historical_corr: pd.DataFrame,
+    ) -> Dict[str, Any]:
+        """Check for correlation breakdown."""
+        
+        # Calculate correlation changes
+        corr_diff = current_corr - historical_corr
+        
+        # Find significant changes
+        significant_changes = []
+        
+        for i in range(len(corr_diff.columns)):
+            for j in range(i + 1, len(corr_diff.columns)):
+                asset1 = corr_diff.columns[i]
+                asset2 = corr_diff.columns[j]
+                change = corr_diff.iloc[i, j]
+                
+                if abs(change) > self.correlation_threshold:
+                    significant_changes.append({
+                        "asset1": asset1,
+                        "asset2": asset2,
+                        "change": change,
+                        "current": current_corr.iloc[i, j],
+                        "historical": historical_corr.iloc[i, j],
+                    })
+        
+        return {
+            "drift_detected": len(significant_changes) > 0,
+            "significant_changes": significant_changes,
+            "threshold": self.correlation_threshold,
+        }
+    
+    def check_strategy_drift(
+        self,
+        recent_returns: pd.Series,
+        historical_returns: pd.Series,
+    ) -> Dict[str, Any]:
+        """Check for strategy performance drift."""
+        
+        # Calculate rolling stats
+        recent_mean = recent_returns.mean() * 252
+        recent_vol = recent_returns.std() * np.sqrt(252)
+        recent_sharpe = recent_mean / recent_vol if recent_vol > 0 else 0
+        
+        historical_mean = historical_returns.mean() * 252
+        historical_vol = historical_returns.std() * np.sqrt(252)
+        historical_sharpe = historical_mean / historical_vol if historical_vol > 0 else 0
+        
+        # Detect drift
+        return_drift = recent_mean - historical_mean
+        vol_drift = recent_vol - historical_vol
+        sharpe_drift = recent_sharpe - historical_sharpe
+        
+        # Significant if Sharpe dropped by more than 0.5
+        drift_detected = sharpe_drift < -0.5
+        
+        return {
+            "drift_detected": drift_detected,
+            "metrics": {
+                "recent_return": recent_mean,
+                "historical_return": historical_mean,
+                "return_drift": return_drift,
+                "recent_vol": recent_vol,
+                "historical_vol": historical_vol,
+                "vol_drift": vol_drift,
+                "recent_sharpe": recent_sharpe,
+                "historical_sharpe": historical_sharpe,
+                "sharpe_drift": sharpe_drift,
+            },
+        }
+
+
+# ============================================================================
+# SCHEDULED RUNNER
+# ============================================================================
+
+class ScheduledRunner:
+    """Run scheduled analysis tasks."""
+    
+    def __init__(self):
+        self.tasks: Dict[str, Dict[str, Any]] = {}
+        self.running = False
+    
+    def add_task(
+        self,
+        task_id: str,
+        func: Callable,
+        schedule: str,  # daily, weekly, monthly
+        time: str = "09:00",  # HH:MM
+        args: tuple = None,
+        kwargs: dict = None,
+    ):
+        """Add a scheduled task."""
+        
+        self.tasks[task_id] = {
+            "func": func,
+            "schedule": schedule,
+            "time": time,
+            "args": args or (),
+            "kwargs": kwargs or {},
+            "last_run": None,
+            "enabled": True,
+        }
+    
+    def remove_task(self, task_id: str):
+        """Remove a scheduled task."""
+        
+        if task_id in self.tasks:
+            del self.tasks[task_id]
+    
+    def enable_task(self, task_id: str, enabled: bool = True):
+        """Enable or disable a task."""
+        
+        if task_id in self.tasks:
+            self.tasks[task_id]["enabled"] = enabled
+    
+    async def run_once(self):
+        """Check and run due tasks once."""
+        
+        now = datetime.now()
+        current_time = now.strftime("%H:%M")
+        
+        for task_id, task in self.tasks.items():
+            if not task["enabled"]:
+                continue
+            
+            # Check if task is due
+            if self._is_task_due(task, now):
+                try:
+                    result = task["func"](*task["args"], **task["kwargs"])
+                    if asyncio.iscoroutine(result):
+                        await result
+                    task["last_run"] = now
+                except Exception as e:
+                    print(f"Task {task_id} failed: {e}")
+    
+    def _is_task_due(self, task: Dict[str, Any], now: datetime) -> bool:
+        """Check if a task is due to run."""
+        
+        schedule = task["schedule"]
+        scheduled_time = task["time"]
+        last_run = task["last_run"]
+        
+        current_time = now.strftime("%H:%M")
+        
+        # Check if time matches (within 1 minute)
+        if current_time != scheduled_time:
+            return False
+        
+        # Check if already run today
+        if last_run and last_run.date() == now.date():
+            return False
+        
+        # Check schedule
+        if schedule == "daily":
+            return True
+        elif schedule == "weekly":
+            # Run on Monday
+            return now.weekday() == 0
+        elif schedule == "monthly":
+            # Run on 1st of month
+            return now.day == 1
+        
+        return False
+    
+    async def run_continuous(self, check_interval: int = 60):
+        """Run continuously, checking for due tasks."""
+        
+        self.running = True
+        
+        while self.running:
+            await self.run_once()
+            await asyncio.sleep(check_interval)
+    
+    def stop(self):
+        """Stop the continuous runner."""
+        
+        self.running = False