from __future__ import annotations
import pandas as pd
import numpy as np


def _atr(high: pd.Series, low: pd.Series, close: pd.Series, period: int) -> pd.Series:
    hl = (high - low).abs()
    hc = (high - close.shift()).abs()
    lc = (low - close.shift()).abs()
    tr = pd.concat([hl, hc, lc], axis=1).max(axis=1)
    return tr.rolling(period, min_periods=period).mean()


def supertrend_series(df: pd.DataFrame, length: int, multiplier: float) -> pd.Series:
    atr = _atr(df["High"], df["Low"], df["Close"], length)
    hl2 = (df["High"] + df["Low"]) / 2
    upper = hl2 + multiplier * atr
    lower = hl2 - multiplier * atr

    trend = pd.Series(index=df.index, dtype=float)
    dir_up = True
    prev_upper = np.nan
    prev_lower = np.nan

    for i in range(len(df)):
        if i == 0 or pd.isna(atr.iat[i]):
            trend.iat[i] = np.nan
            prev_upper = upper.iat[i]
            prev_lower = lower.iat[i]
            continue

        cu = min(upper.iat[i], prev_upper) if dir_up else upper.iat[i]
        cl = max(lower.iat[i], prev_lower) if not dir_up else lower.iat[i]

        if df["Close"].iat[i] > cu:
            dir_up = True
        elif df["Close"].iat[i] < cl:
            dir_up = False

        prev_upper = cu if dir_up else upper.iat[i]
        prev_lower = lower.iat[i] if dir_up else cl
        trend.iat[i] = cl if dir_up else cu

    return trend


def add_supertrends(df: pd.DataFrame, settings: list[tuple[int, float]]) -> pd.DataFrame:
    out = df.copy()
    for length, mult in settings:
        col = f"supertrend_{length}_{mult}"
        out[col] = supertrend_series(out, length, mult)
        out[f"bull_{length}_{mult}"] = (out["Close"] >= out[col]).astype(int)
    return out


def compute_meta_trend(df: pd.DataFrame, settings: list[tuple[int, float]]) -> pd.Series:
    bull_cols = [f"bull_{l}_{m}" for l, m in settings]
    return (df[bull_cols].sum(axis=1) == len(bull_cols)).astype(int)