"""
strategy/rr_signals.py
───────────────────────
RR-v1 — Range Mean Reversion

Logic:
  1. 1H regime filter  : ADX(14) < 18 → market is ranging (not trending)
  2. 15m entry signals : Bollinger Bands(20,2) + RSI(14)
     Long : close < lower BB AND RSI < 30 AND next candle bullish/wick rejection
     Short: close > upper BB AND RSI > 70 AND next candle bearish confirmation
  3. Stop loss : max(recent swing, 1.2 × ATR)
  4. Take profits: TP1 = BB midline (SMA20), TP2 = opposite BB

Extra columns:
  signal       :  1 | -1 | 0
  signal_type  : 'rr_long' | 'rr_short' | ''
  sl_price     : stop loss
  tp1, tp2     : take profit levels
"""

import numpy as np
import pandas as pd

# ─── Parameters ──────────────────────────────────────────────────────────────
ADX_RANGE_MAX   = 18      # ADX must be below this for ranging regime
RSI_OVERSOLD    = 30      # RSI threshold for longs
RSI_OVERBOUGHT  = 70      # RSI threshold for shorts
ATR_SL_MULT     = 1.2     # SL = max(swing, 1.2 × ATR)
SWING_LOOKBACK  = 5       # candles to look back for swing SL


# ─── Main Signal Generator ────────────────────────────────────────────────────

def generate_rr_signals(df: pd.DataFrame) -> pd.DataFrame:
    """
    Scan every 15m candle for RR-v1 mean reversion entries.

    Expects df to have:
      - Standard OHLCV + rsi, atr, adx
      - bb_lower, bb_upper, bb_mid (Bollinger Bands)
      - htf_adx: ADX from 1H timeframe for regime filter
    """
    df = df.copy()
    df['signal']      = 0
    df['signal_type'] = ''
    df['sl_price']    = np.nan
    df['tp1']         = np.nan
    df['tp2']         = np.nan

    required = ['rsi', 'atr', 'bb_lower', 'bb_upper', 'bb_mid', 'htf_adx']

    sig_col   = df.columns.get_loc('signal')
    stype_col = df.columns.get_loc('signal_type')
    sl_col    = df.columns.get_loc('sl_price')
    tp1_col   = df.columns.get_loc('tp1')
    tp2_col   = df.columns.get_loc('tp2')

    for i in range(SWING_LOOKBACK + 1, len(df) - 1):
        row      = df.iloc[i]
        next_row = df.iloc[i + 1]   # confirmation candle

        if row[required].isna().any():
            continue

        # ── Regime filter: only trade when 1H ADX < 18 (ranging) ──────────
        if row['htf_adx'] >= ADX_RANGE_MAX:
            continue

        # ── Long setup ─────────────────────────────────────────────────────
        if (
            row['close']  < row['bb_lower']
            and row['rsi'] < RSI_OVERSOLD
        ):
            # Confirmation: next candle bullish OR shows wick rejection
            bullish_confirm = next_row['close'] > next_row['open']
            wick_rejection  = (next_row['close'] - next_row['low']) > (
                                0.6 * (next_row['high'] - next_row['low']))
            if bullish_confirm or wick_rejection:
                entry = next_row['close']
                # SL = max(recent swing low, 1.2x ATR below entry)
                swing_low = df['low'].iloc[i - SWING_LOOKBACK:i + 1].min()
                sl        = min(swing_low * 0.999, entry - ATR_SL_MULT * row['atr'])
                risk      = entry - sl
                if risk > 0:
                    df.iloc[i + 1, sig_col]   = 1
                    df.iloc[i + 1, stype_col] = 'rr_long'
                    df.iloc[i + 1, sl_col]    = sl
                    df.iloc[i + 1, tp1_col]   = row['bb_mid']     # BB midline
                    df.iloc[i + 1, tp2_col]   = row['bb_upper']   # opposite BB

        # ── Short setup ────────────────────────────────────────────────────
        elif (
            row['close']  > row['bb_upper']
            and row['rsi'] > RSI_OVERBOUGHT
        ):
            bearish_confirm = next_row['close'] < next_row['open']
            wick_rejection  = (next_row['high'] - next_row['close']) > (
                                0.6 * (next_row['high'] - next_row['low']))
            if bearish_confirm or wick_rejection:
                entry = next_row['close']
                swing_high = df['high'].iloc[i - SWING_LOOKBACK:i + 1].max()
                sl         = max(swing_high * 1.001, entry + ATR_SL_MULT * row['atr'])
                risk       = sl - entry
                if risk > 0:
                    df.iloc[i + 1, sig_col]   = -1
                    df.iloc[i + 1, stype_col] = 'rr_short'
                    df.iloc[i + 1, sl_col]    = sl
                    df.iloc[i + 1, tp1_col]   = row['bb_mid']
                    df.iloc[i + 1, tp2_col]   = row['bb_lower']

    return df