import yfinance as yf
import pandas_ta as ta
import pandas as pd
import logging
from typing import List, Dict

logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

DEFAULT_TICKERS = [
    "AAPL", "MSFT", "NVDA", "AMZN", "META", "GOOGL", "TSLA",
    "AMD", "NFLX", "QCOM", "INTC", "BA", "DIS", "SPY", "QQQ"
]

def scan_for_candidates(tickers: List[str] = DEFAULT_TICKERS, min_price: float = 20.0, min_volume: int = 2_000_000) -> pd.DataFrame:
    """
    Scans a list of tickers to find the best candidates for the Touch & Turn strategy.
    Prioritizes high Average True Range (ATR) as a percentage of price, ensuring adequate volume.
    """
    logger.info(f"Scanning {len(tickers)} tickers for high volatility/liquidity...")
    results = []

    # Download daily data for the past 1mo to calculate 14-day ATR and Avg Volume
    data = yf.download(tickers, period="1mo", interval="1d", group_by="ticker", progress=False)

    for ticker in tickers:
        try:
            # Handle single ticker vs multi-ticker dataframe structure from yfinance
            if len(tickers) == 1:
                df = data.copy()
            else:
                df = data[ticker].copy()

            if df.empty or len(df) < 15:
                continue
            
            # Clean column names (yfinance multi-index can sometimes leave tuple names)
            if isinstance(df.columns, pd.MultiIndex):
                df.columns = df.columns.droplevel(1)
                
            df.ta.atr(length=14, append=True)
            
            # Safely get the close price, falling back to yesterday if today's is NaN (common at exactly 09:30)
            close_price = df['Close'].iloc[-1]
            if pd.isna(close_price) and len(df) > 1:
                close_price = df['Close'].iloc[-2]
                
            yesterday_atr = df['ATRr_14'].iloc[-2]
            
            # Safely get avg volume
            avg_volume = df['Volume'].tail(14).mean()
            if pd.isna(avg_volume):
                avg_volume = 0
            
            # Filters
            if pd.isna(close_price) or close_price < min_price or avg_volume < min_volume or pd.isna(yesterday_atr):
                continue
                
            atr_percent = (yesterday_atr / close_price) * 100
            
            results.append({
                "Ticker": ticker,
                "Close": round(float(close_price), 2),
                "ATR_14": round(float(yesterday_atr), 2),
                "ATR_Percent": round(float(atr_percent), 2),
                "Avg_Volume": int(avg_volume)
            })
            
        except Exception as e:
            logger.debug(f"Failed processing {ticker}: {e}")

    results_df = pd.DataFrame(results)
    if not results_df.empty:
        # Sort by ATR Percentage descending (we want the most volatile stocks)
        results_df = results_df.sort_values(by="ATR_Percent", ascending=False).reset_index(drop=True)
        
    return results_df

if __name__ == "__main__":
    candidates = scan_for_candidates()
    print("\nTop Candidates for Touch & Turn Strategy:")
    print("-" * 65)
    print(candidates.to_string(index=False))