ForexBot2/signal_aggregator.py

"""
ForexBot2 — Signal Aggregator
==============================
Combines multi-timeframe technical analysis with optional news context
into a graded actionable signal.
"""

import hashlib
import json
import logging
import os
from dataclasses import asdict, dataclass, field
from datetime import datetime, timezone
from typing import Dict, List, Optional
from zoneinfo import ZoneInfo

import pandas as pd

_NY_TZ = ZoneInfo("America/New_York")

# Prefix of the cost-floor veto both core setup generators emit; the XAU_CP
# cleanup and the post-archetype stale-veto pass match on it, so it is
# spelled once.
_SPREAD_GATE_PREFIX = "SPREAD GATE:"

from chart_patterns import detect_chart_pattern
from config import PlatformConfig, SignalStrength, Timeframe
from providers import BaseDataProvider, make_data_provider
from engine import TechnicalEngine, TechnicalAnalysis, regime_weight_table
from news_engine import NewsEngine, FundamentalContext
import signal_log

logger = logging.getLogger(__name__)


def _is_forex_weekend(now_utc: datetime) -> bool:
    """Gold/FX market closed for the weekend: Friday 17:00 → Sunday 18:00
    NEW YORK time (= Fri 21:00 UTC in US summer, 22:00 in winter —
    matches measured Oanda feed behavior; 2026-08-04 audit I43).
    Deliberate copy of providers.base._is_forex_weekend — a Module 21
    test asserts the copies agree across DST transitions."""
    ny = now_utc.astimezone(_NY_TZ)
    wd = ny.weekday()
    if wd == 4:                      # Friday
        return ny.hour >= 17
    if wd == 5:                      # Saturday
        return True
    if wd == 6:                      # Sunday
        return ny.hour < 18
    return False


@dataclass
class TradeSetup:
    pair: str
    display_name: str
    direction: str  # "BUY" / "SELL"
    entry_price: float
    stop_loss: float
    take_profit_1: float
    take_profit_2: float
    risk_reward: float
    risk_pips: float
    reward_pips: float
    pip_value: float
    # Entry archetype — selects trail params and explains the setup to
    # advisory consumers. setup_pattern is populated only by XAU_CP.
    strategy: str = "GENERIC"
    setup_pattern: Optional[str] = None
    # Strategy cost assumption, not a broker/data-provider quote.
    round_trip_cost: Optional[float] = None
    # Trailing exit params stamped from the PRODUCING config, so per-TF
    # factory overrides (e.g. M15 arm 0.70, M5 dist 0.85) reach every
    # consumer. None = legacy/unstamped setup — formatters fall back to
    # platform defaults; 0.0 means trailing is genuinely disabled.
    trail_arm_r: Optional[float] = None
    trail_dist_r: Optional[float] = None
    # Post-decision stake multiplier from position_sizing (advisory
    # overlay). None = unsized — every gate/exit path ignores it, and
    # the backtester's R accounting is unaffected unless it opts in.
    stake_multiplier: Optional[float] = None


@dataclass
class SignalReport:
    symbol: str
    display_name: str
    timestamp: str

    primary_signal: SignalStrength = SignalStrength.NEUTRAL
    primary_score: float = 0.0
    primary_confidence: float = 0.0

    mtf_agreement: bool = False
    mtf_signals: Dict[str, SignalStrength] = field(default_factory=dict)
    mtf_scores: Dict[str, float] = field(default_factory=dict)

    primary_analysis: Optional[TechnicalAnalysis] = None
    confirmation_analyses: List[TechnicalAnalysis] = field(default_factory=list)

    event_risk: str = "clear"
    event_description: str = ""
    news_sentiment: float = 0.0
    news_description: str = ""
    news_summary: str = ""  # LLM market-context blurb (display-only)

    # LLM agent-panel consensus (advisory/DISPLAY-ONLY). Populated
    # post-hoc by consumers via agent_panel.attach_panel — NEVER inside
    # analyze_from_dataframes (backtest parity) and never read by any
    # gate or scoring path.
    panel_verdict: str = ""        # "AGREE" / "NEUTRAL" / "OPPOSE" / ""
    panel_confidence: float = 0.0
    panel_rationale: str = ""
    panel_takes: List[Dict] = field(default_factory=list)
    panel_created_at: str = ""
    panel_cache_status: str = ""
    panel_payload_hash: str = ""
    panel_verdict_source: str = ""  # "mean" / "macro" / "chair" / ""
    # v2 panel (2026-09-17): the mean persona probability (0-100) that
    # the trade closes profitably, the prompt version that produced it,
    # the exact per-persona payloads and the shared code-computed
    # context — all advisory/DISPLAY-ONLY, logged for the offline replay
    # harness (scripts/panel_replay.py) and the calibration scorecard.
    panel_probability: float = 0.0
    panel_prompt_version: str = ""
    panel_payloads: Dict = field(default_factory=dict)
    panel_context: Dict = field(default_factory=dict)

    # Violent-move watch line (advisory/DISPLAY-ONLY). Stamped post-hoc
    # by surfaces via violent_watch.attach_watch — never populated
    # inside analyze_from_dataframes (backtest parity) and never read by
    # any gate, grade or setup generator.
    watch_line: str = ""

    final_signal: SignalStrength = SignalStrength.NEUTRAL
    final_score: float = 0.0
    final_confidence: float = 0.0
    action_recommendation: str = ""
    signal_quality: str = ""  # A+, A, B, C, D, F

    trade_setup: Optional[TradeSetup] = None

    bullish_reasons: List[str] = field(default_factory=list)
    bearish_reasons: List[str] = field(default_factory=list)
    warnings: List[str] = field(default_factory=list)
    # Warnings a shipped setup superseded, removed from `warnings` but
    # logged so they stay auditable (metadata only): the vetoes the
    # XAU_CP acceptance stripped and, since 2026-09-08, a core
    # generator's stale spread-gate veto when a later generator produced
    # the setup (`_drop_stale_spread_gate_vetoes`).
    cleaned_warnings: List[str] = field(default_factory=list)

    current_price: float = 0.0
    spread_estimate: float = 0.0
    market_regime: str = ""
    volatility_regime: str = ""
    exhaustion_count: int = 0
    # Record-keeping only (no gate reads these): the penalty tags that
    # fired in _compute_final_signal (tag -> RAW pre-decay amount), the
    # decayed total actually subtracted from confidence, and whether the
    # momentum-fading half of the exhaustion block was true.
    penalty_tags: Dict[str, float] = field(default_factory=dict)
    penalty_total: float = 0.0
    momentum_fading: bool = False


class SignalAggregator:
    """Master aggregator for live signal analysis."""

    def __init__(self, config: PlatformConfig):
        self.config = config
        self.data_provider = make_data_provider(config)
        self.tech_engine = TechnicalEngine(config.technical)
        self._news_engine: Optional[NewsEngine] = None  # lazy

    def _get_news_engine(self) -> NewsEngine:
        if self._news_engine is None:
            self._news_engine = NewsEngine(self.config.news)
        return self._news_engine

    # ── Public: live mode entry point ──────────────────────────────

    def analyze_pair(
        self,
        symbol: Optional[str] = None,
        log_source: Optional[str] = None,
    ) -> SignalReport:
        """Live-mode entry: fetch all needed timeframes then delegate.

        Every report produced here (including weekend/day-filter early
        returns) is persisted to the signal-history DB via _log_signal —
        fail-safe, and never reached by the backtester, which calls
        analyze_from_dataframes directly (live == backtest parity)."""
        if symbol is None:
            symbol = self.config.symbol

        pair_info = self.data_provider.get_pair_info(symbol)
        now_utc = datetime.now(timezone.utc)

        if pair_info["display"] == "XAU/USD" and self.config.xau_strategy.enabled:
            xau_cfg = self.config.xau_strategy

            if xau_cfg.weekend_filter and _is_forex_weekend(now_utc):
                report = SignalReport(
                    symbol=symbol,
                    display_name=pair_info["display"],
                    timestamp=str(now_utc),
                )
                report.final_signal = SignalStrength.NEUTRAL
                report.signal_quality = "F"
                report.trade_setup = None
                report.warnings.append("WEEKEND FILTER: XAU/USD market closed — no liquidity")
                report.action_recommendation = "NO TRADE — Weekend (market closed)"
                self._log_signal(report, log_source, blocked=True)
                return report

            if xau_cfg.allowed_weekdays and now_utc.weekday() not in xau_cfg.allowed_weekdays:
                day_names = {0: "Mon", 1: "Tue", 2: "Wed", 3: "Thu", 4: "Fri", 5: "Sat", 6: "Sun"}
                allowed = "/".join(day_names.get(d, str(d)) for d in xau_cfg.allowed_weekdays)
                report = SignalReport(
                    symbol=symbol,
                    display_name=pair_info["display"],
                    timestamp=str(now_utc),
                )
                report.final_signal = SignalStrength.NEUTRAL
                report.signal_quality = "F"
                report.trade_setup = None
                report.warnings.append(
                    f"DAY FILTER: {day_names.get(now_utc.weekday(), '?')} not in allowed days ({allowed})"
                )
                report.action_recommendation = f"NO TRADE — Only trading {allowed}"
                self._log_signal(report, log_source, blocked=True)
                return report

        timeframes = [self.config.primary_timeframe] + list(
            self.config.confirmation_timeframes
        )

        bt = self.config.backtest
        if bt.regime_filter and bt.regime_filter_timeframe not in timeframes:
            timeframes.append(bt.regime_filter_timeframe)

        dfs: Dict[Timeframe, pd.DataFrame] = {}
        stale_tfs = []
        for tf in timeframes:
            try:
                df = self.data_provider.fetch_ohlcv(symbol, tf)
                dfs[tf] = df
                if not self.data_provider.is_data_fresh(df, symbol, tf):
                    stale_tfs.append(tf.value)
            except Exception as e:
                logger.warning(f"Fetch failed for {symbol} {tf.value}: {e}")

        # A missing frame is a FAILED scan, never a degraded one. Losing a
        # CONFIRMATION frame used to let the scan continue with fewer
        # frames: _check_mtf_agreement needs >= 2 signals, so it returned
        # False and the composite took a 15-point mtf_disagreement penalty
        # the market never earned (A/A+ became impossible), while on
        # M15/H1/H4 a B/C setup could still ship on one confirmation with
        # no warning at all. The backtester never sees that state
        # (run_backtest raises on an empty frame), so it was a live-only
        # divergence. 2026-09-13 audit.
        missing = [tf for tf in timeframes if tf not in dfs]
        if missing:
            report = SignalReport(
                symbol=symbol,
                display_name=pair_info["display"],
                timestamp=str(now_utc),
            )
            report.final_signal = SignalStrength.NEUTRAL
            report.signal_quality = "F"
            report.trade_setup = None
            if self.config.primary_timeframe in missing:
                # Legacy text signal_log.gate_reason already recognises.
                report.warnings.append("Primary timeframe data unavailable")
            report.warnings.append(
                f"DATA UNAVAILABLE: {', '.join(tf.value for tf in missing)}"
                " — scan failed"
            )
            report.action_recommendation = "NO TRADE — market data unavailable"
            self._log_signal(report, log_source, dfs, failed=True)
            return report

        report = self.analyze_from_dataframes(
            symbol=symbol,
            dfs=dfs,
            use_news=True,
        )
        # Live-only hand-off for the post-hoc advisory add-ons (the agent
        # panel's price context reads the frames this scan analysed, so
        # it sees exactly the bars the engine saw). A private attribute
        # like _signal_log_rowid: not a dataclass field, never read by
        # the core, absent on backtest replays by construction.
        report._frames = dfs

        if stale_tfs:
            report.warnings.append(f"STALE DATA on {', '.join(stale_tfs)} — signal may be unreliable")
            if self.config.primary_timeframe.value in stale_tfs:
                report.final_signal = SignalStrength.NEUTRAL
                report.signal_quality = "F"
                report.trade_setup = None
                report.warnings.append("Primary timeframe stale — signal suppressed")

        if bt.regime_filter:
            self._apply_regime_filter_live(report, dfs)

        # Session window + blocked hours on the signal bar's CLOSE time —
        # the exact predicate the backtester's replay loop applies before
        # each bar, so live and backtest cannot diverge. Bar-close time
        # (not wall clock) makes manual mid-candle scans behave like
        # boundary auto-scans. This replaces an older wall-clock filter
        # that was guarded by a marker string no code wrote.
        if bt.session_filter and report.primary_analysis is not None:
            try:
                bar_close = (
                    pd.Timestamp(report.timestamp)
                    + pd.Timedelta(
                        seconds=BaseDataProvider.CANDLE_SECONDS[
                            self.config.primary_timeframe
                        ]
                    )
                )
                close_h = bar_close.hour
                _is_xau = pair_info["display"] == "XAU/USD"
                _blocked = (self.config.xau_strategy.blocked_hours_utc
                            if _is_xau else bt.blocked_hours_utc)
                outside = not (bt.session_start_hour <= close_h < bt.session_end_hour)
                if outside or close_h in _blocked:
                    report.warnings.append(
                        f"SESSION FILTER: bar closes {close_h:02}:00 UTC — "
                        + ("outside window" if outside else "blocked hour")
                    )
                    report.final_signal = SignalStrength.NEUTRAL
                    report.signal_quality = "F"
                    report.action_recommendation = "NO TRADE — Outside tradeable session"
                    report.trade_setup = None
            except Exception:
                # Fail-open by design (an unparseable timestamp shouldn't
                # kill the scan) but never silently: this filter guards
                # real money hours.
                logger.exception(
                    f"{symbol}: live session filter errored — filter SKIPPED this scan"
                )

        self._log_signal(report, log_source, dfs)
        return report

    def _log_signal(
        self, report: SignalReport, source: Optional[str],
        dataframes: Optional[Dict[Timeframe, pd.DataFrame]] = None,
        blocked: bool = False, failed: bool = False,
    ) -> None:
        """Persist the scan to the signal-history DB (signal_log.py).

        Fail-safe by contract — a logging failure must never break a
        live scan. The news context re-fetch rides the NewsEngine's
        class-level caches, so it costs nothing within a scan and
        matches the context the signal just used; if it fails, the row
        is still written, just without the news JSON blobs. ``blocked``
        marks weekend/day early returns: no news fetch, no snapshots —
        a ~1 KB liveness row (or nothing, per signal_log.log_blocked_scans).
        A scan missing ANY required frame (primary or confirmation, or
        the regime-filter frame when that filter is on) is logged as
        ``scan_kind`` ``failed`` (price 0, no signal): liveness evidence
        that readers skip like ``blocked``, so it never pollutes NEUTRAL
        statistics."""
        context = None
        if not blocked:
            try:
                context = self._get_news_engine().get_fundamental_context()
            except Exception:
                logger.debug("signal log: news context unavailable",
                             exc_info=True)
        kind = "blocked" if blocked else "scan"
        primary = (dataframes or {}).get(self.config.primary_timeframe)
        if failed or (not blocked and dataframes is not None
                      and (primary is None or primary.empty)):
            kind = "failed"
        extra = {"scan_kind": kind}
        for key, fingerprint in (
                ("config_fingerprint", self.gate_config_fingerprint),
                ("engine_fingerprint", self.engine_config_fingerprint)):
            try:
                extra[key] = fingerprint()
            except Exception:
                logger.debug("signal log: %s unavailable", key, exc_info=True)
        signal_log.log_signal(
            report,
            context=context,
            source=source or "live",
            timeframe=self.config.primary_timeframe.value,
            dataframes=dataframes,
            extra=extra,
        )

    def gate_config_fingerprint(self) -> str:
        """Short hash of the gate configuration this process runs with.

        Logged on every row so two surfaces disagreeing on the same bar
        (2026-08-19: the dashboard process applied a different blocked-
        hours list than the bot for a day) is diagnosable from the DB."""
        xs = self.config.xau_strategy
        bt = self.config.backtest
        payload = {
            "primary": self.config.primary_timeframe.value,
            "confirm": [tf.value for tf in self.config.confirmation_timeframes],
            "blocked_hours": list(xs.blocked_hours_utc),
            "sell_blocked_hours": list(xs.sell_blocked_hours_utc),
            "allowed_weekdays": list(xs.allowed_weekdays),
            "blocked_regimes": list(xs.blocked_regimes),
            "sell_blocked_regimes": list(xs.sell_blocked_regimes),
            "sell_min_quality": list(xs.sell_min_quality),
            "sell_vol_regimes": list(xs.sell_vol_regimes),
            "big_candle": xs.big_candle_atr_multiple,
            "cp": bool(xs.cp_enabled),
            "pb": bool(xs.m5_pullback_enabled),
            "session": [bt.session_filter, bt.session_start_hour,
                        bt.session_end_hour],
            "regime_filter": bool(bt.regime_filter),
        }
        return hashlib.sha1(
            json.dumps(payload, sort_keys=True, default=str).encode("utf-8")
        ).hexdigest()[:12]

    def engine_config_fingerprint(self) -> str:
        """Short hash of the SCORING configuration this process runs with.

        gate_config_fingerprint covers the post-grade gates; this covers
        what produced the composite in the first place — indicator
        parameters, signal thresholds/geometry, the regime weight
        multipliers actually in force (override included) and the
        timeframe set. Logged on every row so indicator-weight research
        can tell which rows share a scoring configuration."""
        payload = {
            "technical": asdict(self.config.technical),
            "signals": asdict(self.config.signals),
            "regime_weights": regime_weight_table(self.config.technical),
            "primary": self.config.primary_timeframe.value,
            "confirm": [tf.value for tf in self.config.confirmation_timeframes],
            "display_name": self.config.display_name,
        }
        return hashlib.sha1(
            json.dumps(payload, sort_keys=True, default=str).encode("utf-8")
        ).hexdigest()[:12]

    def _apply_regime_filter_live(
        self, report: SignalReport, dfs: Dict[Timeframe, pd.DataFrame]
    ) -> None:
        """If higher-TF ADX is below threshold, downgrade signal to NEUTRAL."""
        bt = self.config.backtest
        regime_tf = bt.regime_filter_timeframe
        df = dfs.get(regime_tf)
        if df is None or df.empty:
            report.warnings.append(
                f"Regime filter requested but {regime_tf.value} data unavailable"
            )
            return

        adx_value = self._latest_adx(df, self.config.technical.adx_period)
        if adx_value is None:
            report.warnings.append(
                f"Regime filter: ADX could not be computed on {regime_tf.value}"
            )
            return

        if adx_value < bt.regime_filter_min_adx:
            report.warnings.append(
                f"REGIME FILTER: {regime_tf.value} ADX={adx_value:.1f} < "
                f"{bt.regime_filter_min_adx:.0f} — market choppy, skipping signal"
            )
            report.final_signal = SignalStrength.NEUTRAL
            report.signal_quality = "F"
            report.action_recommendation = (
                f"NO TRADE — Regime filter (ADX {adx_value:.1f} on {regime_tf.value})"
            )
            report.trade_setup = None
        else:
            report.bullish_reasons.append(
                f"Regime: {regime_tf.value} ADX={adx_value:.1f} confirms trending market"
            ) if report.final_score > 0 else report.bearish_reasons.append(
                f"Regime: {regime_tf.value} ADX={adx_value:.1f} confirms trending market"
            )

    @staticmethod
    def _latest_adx(df: pd.DataFrame, period: int = 14) -> Optional[float]:
        """Wilder ADX on a fresh df; returns the last finite value or None."""
        if df is None or len(df) < period * 3:
            return None
        high, low, close = df["High"], df["Low"], df["Close"]
        tr = pd.concat([
            high - low,
            (high - close.shift()).abs(),
            (low - close.shift()).abs(),
        ], axis=1).max(axis=1)
        plus_dm_raw = high.diff()
        minus_dm_raw = -low.diff()
        plus_dm = plus_dm_raw.where((plus_dm_raw > minus_dm_raw) & (plus_dm_raw > 0), 0.0)
        minus_dm = minus_dm_raw.where((minus_dm_raw > plus_dm_raw) & (minus_dm_raw > 0), 0.0)
        atr_s = tr.ewm(alpha=1/period, min_periods=period, adjust=False).mean()
        plus_di = 100 * (plus_dm.ewm(alpha=1/period, min_periods=period, adjust=False).mean() / atr_s)
        minus_di = 100 * (minus_dm.ewm(alpha=1/period, min_periods=period, adjust=False).mean() / atr_s)
        denom = (plus_di + minus_di).replace(0, float("nan"))
        dx = 100 * (plus_di - minus_di).abs() / denom
        adx = dx.ewm(alpha=1/period, min_periods=period, adjust=False).mean().dropna()
        if adx.empty:
            return None
        return float(adx.iloc[-1])

    def _detect_big_candle(self, primary_df: pd.DataFrame) -> Optional[tuple]:
        """Largest abnormal candle in the recent primary-TF window.

        Returns (timestamp, ratio) of the biggest bar whose range is
        >= big_candle_atr_multiple x the PRIOR bar's ATR within the last
        big_candle_lookback_hours of primary-TF bars, else None. The
        prior-bar ATR (shift 1) matters: a huge candle inflates the
        Wilder ATR immediately, so comparing against the concurrent ATR
        under-detects the very bars this gate exists to catch.

        Pure pandas on the caller-sliced df, so the live path and the
        backtest replay (which calls analyze_from_dataframes on prefix
        slices) compute bit-identical answers. The tail slice keeps the
        per-call cost bounded (~n + 42 rows); min_periods NaNs and short
        histories degrade to "no trip".
        """
        cfg = self.config.xau_strategy
        bar_secs = BaseDataProvider.CANDLE_SECONDS[self.config.primary_timeframe]
        n = max(1, int(cfg.big_candle_lookback_hours * 3600 // bar_secs))
        period = self.config.technical.atr_period
        tail = primary_df.iloc[-(n + 3 * period):]
        high, low, close = tail["High"], tail["Low"], tail["Close"]
        tr = pd.concat(
            [high - low,
             (high - close.shift()).abs(),
             (low - close.shift()).abs()],
            axis=1,
        ).max(axis=1)
        atr = tr.ewm(alpha=1 / period, min_periods=period, adjust=False).mean()
        ratio = (high - low) / atr.shift(1)
        window = ratio.iloc[-n:]
        if window.empty:
            return None
        max_ratio = window.max()
        if pd.isna(max_ratio) or max_ratio < cfg.big_candle_atr_multiple:
            return None
        return window.idxmax(), float(max_ratio)

    def _apply_m5_filters(self, report: SignalReport) -> None:
        """Extra selectivity for M5 — noisy TF needs stronger confirmation."""
        if report.signal_quality not in ("A+", "A", "B", "C"):
            return

        xau_cfg = self.config.xau_strategy
        abs_score = abs(report.final_score)

        # Gate 1: minimum absolute score (M5 composite scores near the
        # threshold are noise — require a wider margin).
        if abs_score < xau_cfg.m5_min_abs_score:
            report.warnings.append(
                f"M5 FILTER: |score| {abs_score:.1f} < {xau_cfg.m5_min_abs_score:.0f} minimum"
            )
            report.signal_quality = "D"
            report.trade_setup = None
            return

        # Gate 2: require MTF agreement (M5 alone is too noisy).
        if xau_cfg.m5_require_mtf and not report.mtf_agreement:
            report.warnings.append(
                "M5 FILTER: no MTF agreement — M5 signals require confirmation"
            )
            report.signal_quality = "D"
            report.trade_setup = None
            return

        # Gate 3: at least one confirmation TF must have a meaningful
        # score in the same direction — NEUTRAL confirmations don't
        # count for M5 since they indicate the move isn't visible on
        # higher timeframes yet.
        min_conf_score = xau_cfg.m5_min_confirmation_score
        is_buy = report.final_score > 0
        confirmed = False
        primary_tf = self.config.primary_timeframe.value
        for tf_val, tf_score in report.mtf_scores.items():
            if tf_val == primary_tf:
                continue
            if is_buy and tf_score >= min_conf_score:
                confirmed = True
                break
            if not is_buy and tf_score <= -min_conf_score:
                confirmed = True
                break

        if not confirmed:
            report.warnings.append(
                "M5 FILTER: no confirmation TF shows directional conviction"
            )
            report.signal_quality = "D"
            report.trade_setup = None
            return

        # Gate 4: cap SELL quality on M5 — A-grade SELLs enter at the
        # extreme of temporary dips in gold's macro uptrend and lose
        # money. B-grade SELLs (weaker conviction, shallower entries)
        # are marginally profitable.
        is_sell = report.final_score < 0
        if is_sell and report.signal_quality not in xau_cfg.m5_sell_max_quality:
            report.warnings.append(
                f"M5 FILTER: SELL quality {report.signal_quality} capped — "
                f"high-conviction M5 shorts enter at extremes"
            )
            report.signal_quality = "D"
            report.trade_setup = None
            return

        # Gate 5: block chop-transition volatility. 'high' vol on M5 is
        # elevated noise without the directional expansion of 'extreme'
        # — stops get swept before targets (61% WR, negative total in
        # every backtest window; low/normal/extreme all positive).
        if report.volatility_regime in xau_cfg.m5_blocked_vol_regimes:
            report.warnings.append(
                f"M5 FILTER: '{report.volatility_regime}' volatility — "
                f"chop-transition regime, stops get swept"
            )
            report.signal_quality = "D"
            report.trade_setup = None
            return

        # Gate 6: SELL RSI floor — shorting with RSI already < 25 means
        # entering at the extreme of the dip (same pathology as Gate 4,
        # visible through RSI instead of quality grade).
        if is_sell:
            rsi_ind = report.primary_analysis.indicators.get("rsi")
            if rsi_ind is not None and rsi_ind.value < xau_cfg.m5_sell_min_rsi:
                report.warnings.append(
                    f"M5 FILTER: SELL blocked — RSI {rsi_ind.value:.0f} < "
                    f"{xau_cfg.m5_sell_min_rsi:.0f}, dip already extended"
                )
                report.signal_quality = "D"
                report.trade_setup = None
                return

    # ── M5 pullback entry (XAU_PB) ──────────────────────────────────

    def _detect_m5_pullback(
        self, report: SignalReport, primary_df: pd.DataFrame
    ) -> Optional[Dict]:
        """Detect a pullback-continuation candidate on the primary TF.

        Trend intact (EMA alignment >= 3/4, ADX floor) + price retraced
        to EMA21/KC-mid within the last few bars + resumption bar closed
        back in trend direction + RSI in the pullback zone (not the
        extended zone breakout entries fire in).
        """
        cfg = self.config.xau_strategy
        pa = report.primary_analysis
        if pa is None or len(primary_df) < 60:
            return None
        levels = pa.key_levels
        atr = levels.get("atr", 0.0)
        ema_f = levels.get("ema_fast")
        ema_m = levels.get("ema_medium")
        ema_s = levels.get("ema_slow")
        ema_t = levels.get("ema_trend")
        if not atr or None in (ema_f, ema_m, ema_s, ema_t):
            return None

        adx_ind = pa.indicators.get("adx")
        rsi_ind = pa.indicators.get("rsi")
        if adx_ind is None or rsi_ind is None:
            return None
        if adx_ind.value < cfg.m5_pullback_adx_min:
            return None

        close = float(primary_df["Close"].iloc[-1])
        open_ = float(primary_df["Open"].iloc[-1])
        n_touch = cfg.m5_pullback_touch_bars
        recent_low = float(primary_df["Low"].iloc[-n_touch:].min())
        recent_high = float(primary_df["High"].iloc[-n_touch:].max())
        # Pullback swing anchor for SL: extreme of the last 8 bars.
        pb_low = float(primary_df["Low"].iloc[-8:].min())
        pb_high = float(primary_df["High"].iloc[-8:].max())
        kc_mid = levels.get("kc_mid", ema_m)
        lo_atr, hi_atr = cfg.m5_pullback_retrace_atr
        rsi = rsi_ind.value

        # BUY: uptrend structure, dip to the band, bullish resumption.
        up_align = sum([ema_f > ema_m, ema_m > ema_s, ema_s > ema_t,
                        close > ema_t])
        if up_align >= 3:
            touched = recent_low <= max(ema_m, kc_mid)
            swing_high = float(primary_df["High"].iloc[-40:].max())
            depth = (swing_high - pb_low) / atr
            resumed = close > open_ and close > ema_f
            lo_z, hi_z = cfg.m5_pullback_rsi_buy
            if (touched and resumed and lo_atr <= depth <= hi_atr
                    and lo_z <= rsi <= hi_z):
                return {"direction": "BUY", "anchor": pb_low,
                        "retrace_atr": depth}

        # SELL: mirrored.
        dn_align = sum([ema_f < ema_m, ema_m < ema_s, ema_s < ema_t,
                        close < ema_t])
        if dn_align >= 3:
            touched = recent_high >= min(ema_m, kc_mid)
            swing_low = float(primary_df["Low"].iloc[-40:].min())
            depth = (pb_high - swing_low) / atr
            resumed = close < open_ and close < ema_f
            lo_z, hi_z = cfg.m5_pullback_rsi_sell
            if (touched and resumed and lo_atr <= depth <= hi_atr
                    and lo_z <= rsi <= hi_z):
                return {"direction": "SELL", "anchor": pb_high,
                        "retrace_atr": depth}
        return None

    def _try_m5_pullback(
        self, report: SignalReport, primary_df: pd.DataFrame, pip_value: float
    ) -> None:
        """Build an XAU_PB trade setup if a pullback candidate passes gates."""
        cfg = self.config.xau_strategy

        pb = self._detect_m5_pullback(report, primary_df)
        if pb is None:
            return
        direction = pb["direction"]
        is_buy = direction == "BUY"

        # Directional composite floor (lower than Gate 1 by design —
        # price mid-pullback scores weaker on the momentum indicators).
        signed = report.final_score if is_buy else -report.final_score
        if signed < cfg.m5_pullback_min_score:
            return

        # Confirmation TF must agree (reuse Gate-3 threshold).
        min_conf = cfg.m5_min_confirmation_score
        confirmed = False
        primary_tf = self.config.primary_timeframe.value
        for tf_val, tf_score in report.mtf_scores.items():
            if tf_val == primary_tf:
                continue
            if (is_buy and tf_score >= min_conf) or \
               (not is_buy and tf_score <= -min_conf):
                confirmed = True
                break
        if not confirmed:
            return

        # Never against multi-TF consensus (mirror of gate 6f).
        sigs = list(report.mtf_signals.values())
        bull_n = sum(1 for s in sigs if s in (SignalStrength.BUY, SignalStrength.STRONG_BUY))
        bear_n = sum(1 for s in sigs if s in (SignalStrength.SELL, SignalStrength.STRONG_SELL))
        thr = self.config.signals.mtf_agreement_threshold
        if (is_buy and bear_n >= thr) or (not is_buy and bull_n >= thr):
            return

        # Volatility gates match the breakout path.
        if report.volatility_regime in cfg.m5_blocked_vol_regimes:
            return
        if (not is_buy and report.volatility_regime
                and report.volatility_regime not in cfg.sell_vol_regimes):
            return

        atr = report.primary_analysis.key_levels.get("atr", 0.0)
        entry = report.current_price
        min_sl_price = self.config.signals.min_sl_pips * pip_value
        if is_buy:
            sl_dist = max(entry - (pb["anchor"] - 0.5 * atr), min_sl_price)
            sl = entry - sl_dist
            tp1 = entry + 1.5 * sl_dist
            tp2 = entry + 2.5 * sl_dist
        else:
            sl_dist = max((pb["anchor"] + 0.5 * atr) - entry, min_sl_price)
            sl = entry + sl_dist
            tp1 = entry - 1.5 * sl_dist
            tp2 = entry - 2.5 * sl_dist

        # Spread gate — same economics as the breakout generators.
        min_tp = (self.config.signals.round_trip_cost
                  * self.config.signals.min_tp_cost_multiple)
        if 1.5 * sl_dist < min_tp:
            return

        report.trade_setup = TradeSetup(
            pair=report.symbol,
            display_name=report.display_name,
            direction=direction,
            entry_price=entry,
            stop_loss=sl,
            take_profit_1=tp1,
            take_profit_2=tp2,
            risk_reward=1.5,
            risk_pips=sl_dist / pip_value,
            reward_pips=1.5 * sl_dist / pip_value,
            pip_value=pip_value,
            strategy="XAU_PB",
            round_trip_cost=self.config.signals.round_trip_cost,
            trail_arm_r=self.config.signals.pb_trail_arm_r,
            trail_dist_r=self.config.signals.pb_trail_dist_r,
        )
        if report.signal_quality not in ("A+", "A", "B"):
            report.signal_quality = "B"
        report.final_signal = (SignalStrength.BUY if is_buy
                               else SignalStrength.SELL)
        report.action_recommendation = (
            f"{direction} — pullback continuation entry"
        )
        marker = (f"XAU PULLBACK STRATEGY: retrace {pb['retrace_atr']:.1f} "
                  f"ATR to EMA21/KC-mid, trend resumed")
        target = report.bullish_reasons if is_buy else report.bearish_reasons
        target.insert(0, marker)

    # ── Chart-pattern entry (XAU_CP) ────────────────────────────────

    def _try_chart_pattern(
        self, report: SignalReport, primary_df: pd.DataFrame, pip_value: float
    ) -> None:
        """Build an XAU_CP trade setup if a completed chart pattern
        passes the pair-level protections.

        Unlike the composite paths there is no minimum-score gate: a
        reversal pattern completes exactly where the composite is still
        pointing the old way. Protection comes from the pattern's own
        structural gates plus the validated directional filters below.
        """
        cfg = self.config.xau_strategy

        pa = report.primary_analysis
        if pa is None:
            return
        atr = pa.key_levels.get("atr", 0.0)
        # Same $-ATR floor the swing generator applies (dead tape).
        if atr < cfg.min_atr_threshold:
            return

        pat = detect_chart_pattern(primary_df, atr, cfg)
        if pat is None:
            return
        direction = pat["direction"]
        is_buy = direction == "BUY"

        # Never against multi-TF consensus (mirror of gate 6f).
        sigs = list(report.mtf_signals.values())
        bull_n = sum(1 for s in sigs if s in (SignalStrength.BUY, SignalStrength.STRONG_BUY))
        bear_n = sum(1 for s in sigs if s in (SignalStrength.SELL, SignalStrength.STRONG_SELL))
        thr = self.config.signals.mtf_agreement_threshold
        if (is_buy and bear_n >= thr) or (not is_buy and bull_n >= thr):
            return

        # Directional protections — gold's macro-uptrend asymmetry is
        # structural, so the validated SELL gates apply to pattern
        # shorts too (quiet-vol shorts fade, blocked SELL hours lose).
        if not is_buy:
            if (report.volatility_regime
                    and report.volatility_regime not in cfg.sell_vol_regimes):
                return
            if cfg.sell_blocked_hours_utc:
                try:
                    close_h = (
                        pd.Timestamp(report.timestamp)
                        + pd.Timedelta(
                            seconds=BaseDataProvider.CANDLE_SECONDS[
                                self.config.primary_timeframe
                            ]
                        )
                    ).hour
                except Exception:
                    close_h = None
                    logger.warning(
                        f"{report.symbol}: bar-close hour underivable — "
                        f"CP SELL hour gate FAILS CLOSED (short rejected)"
                    )
                if close_h is None or close_h in cfg.sell_blocked_hours_utc:
                    return
            if self.config.primary_timeframe == Timeframe.M5:
                rsi_ind = pa.indicators.get("rsi")
                if rsi_ind is not None and rsi_ind.value < cfg.m5_sell_min_rsi:
                    return

        # M5 chop-transition volatility block (validated, both ways).
        if (self.config.primary_timeframe == Timeframe.M5
                and report.volatility_regime in cfg.m5_blocked_vol_regimes):
            return

        entry = report.current_price
        min_sl_price = self.config.signals.min_sl_pips * pip_value
        height = pat["height"]
        if is_buy:
            sl_dist = max(entry - (pat["anchor"] - 0.5 * atr), min_sl_price)
            sl = entry - sl_dist
            tp1 = entry + height
            tp2 = entry + 1.5 * height
        else:
            sl_dist = max((pat["anchor"] + 0.5 * atr) - entry, min_sl_price)
            sl = entry + sl_dist
            tp1 = entry - height
            tp2 = entry - 1.5 * height

        # The measured move IS the pattern target — if it can't clear
        # the RR floor or the spread gate, the pattern is untradeable.
        rr = height / sl_dist if sl_dist > 0 else 0.0
        if rr < self.config.signals.min_rr_ratio - 1e-9:
            return
        min_tp = (self.config.signals.round_trip_cost
                  * self.config.signals.min_tp_cost_multiple)
        if height < min_tp:
            return

        report.trade_setup = TradeSetup(
            pair=report.symbol,
            display_name=report.display_name,
            direction=direction,
            entry_price=entry,
            stop_loss=sl,
            take_profit_1=tp1,
            take_profit_2=tp2,
            risk_reward=round(rr, 2),
            risk_pips=sl_dist / pip_value,
            reward_pips=height / pip_value,
            pip_value=pip_value,
            strategy="XAU_CP",
            setup_pattern=pat["pattern"],
            round_trip_cost=self.config.signals.round_trip_cost,
            trail_arm_r=self.config.signals.trail_arm_r,
            trail_dist_r=self.config.signals.trail_dist_r,
        )
        if report.signal_quality not in ("A+", "A", "B"):
            report.signal_quality = "B"
        report.final_signal = (SignalStrength.BUY if is_buy
                               else SignalStrength.SELL)
        report.action_recommendation = (
            f"{direction} — {pat['pattern']} pattern completion"
        )
        marker = (f"XAU CHART PATTERN: {pat['pattern']} "
                  f"({direction}) break of {pat['trigger']:.2f}, "
                  f"height {height / atr:.1f} ATR")
        target = report.bullish_reasons if is_buy else report.bearish_reasons
        target.insert(0, marker)
        self._clean_accepted_cp_warnings(report)

    @staticmethod
    def _clean_accepted_cp_warnings(report: SignalReport) -> None:
        """Drop superseded composite-path warnings after XAU_CP accepts.

        This is metadata-only. Pattern detection, gates, confidence,
        exhaustion state, setup geometry, and signal selection are already
        final when this runs.
        """
        stale_prefixes = (
            _SPREAD_GATE_PREFIX,
            "EXHAUSTION BLOCK",
            "EXHAUSTED TREND FILTER:",
            "MTF OPPOSITION:",
            "SELL FILTER: ",
            "M5 FILTER: |score|",
            "M5 FILTER: no MTF agreement",
            "M5 FILTER: no confirmation TF",
            "M5 FILTER: SELL quality",
            "M5 FILTER: SELL blocked",
        )
        extension_prefixes = (
            "TREND EXHAUSTION",
            "NO PULLBACK",
            "FAR FROM EMA200",
            "Extended — RSI",
            "Weak pullback",
        )
        filtered: List[str] = []
        dropped: List[str] = []
        for warning in report.warnings:
            if warning.startswith(stale_prefixes):
                dropped.append(warning)
                continue
            if (warning.startswith("REGIME FILTER: '")
                    and warning.endswith("blocked for XAU")):
                dropped.append(warning)
                continue
            if (warning.startswith("M5 FILTER: '")
                    and "volatility" in warning):
                dropped.append(warning)
                continue
            filtered.append(warning)

        extension_items = [
            (index, warning)
            for index, warning in enumerate(filtered)
            if warning.startswith(extension_prefixes)
        ]
        if not extension_items:
            report.warnings[:] = filtered
            report.cleaned_warnings.extend(dropped)
            return
        first_index = extension_items[0][0]
        priority = {prefix: rank for rank, prefix in enumerate(extension_prefixes)}
        strongest = min(
            (warning for _, warning in extension_items),
            key=lambda warning: next(
                rank for prefix, rank in priority.items()
                if warning.startswith(prefix)),
        )
        kept = [
            warning for warning in filtered
            if not warning.startswith(extension_prefixes)
        ]
        kept.insert(first_index, strongest)
        dropped.extend(
            warning for _, warning in extension_items if warning != strongest)
        report.warnings[:] = kept
        report.cleaned_warnings.extend(dropped)

    # ── Public: bar-replayable entry point ─────────────────────────

    def analyze_from_dataframes(
        self,
        symbol: str,
        dfs: Dict[Timeframe, pd.DataFrame],
        context: Optional[FundamentalContext] = None,
        use_news: bool = True,
    ) -> SignalReport:
        """
        Run the full analysis pipeline on already-fetched dataframes.

        `dfs` must include the primary timeframe and ideally the
        confirmation timeframes. Each df is treated as the visible history
        at decision time — the backtester is responsible for slicing so
        no future bars leak in.
        """
        pair_info = self.data_provider.get_pair_info(symbol)
        display = pair_info["display"]
        pip_value = pair_info["pip_value"]

        report = SignalReport(
            symbol=symbol,
            display_name=display,
            timestamp=str(datetime.now(timezone.utc)),
        )

        # ── 1. Primary timeframe ───────────────────────────────────
        primary_df = dfs.get(self.config.primary_timeframe)
        if primary_df is None or primary_df.empty:
            report.warnings.append("Primary timeframe data unavailable")
            report.signal_quality = "F"
            return report

        primary_analysis = self.tech_engine.analyze(
            primary_df, symbol, self.config.primary_timeframe.value
        )
        report.primary_analysis = primary_analysis
        report.primary_signal = primary_analysis.composite_signal
        report.primary_score = primary_analysis.composite_score
        report.primary_confidence = primary_analysis.confidence
        report.market_regime = primary_analysis.market_regime
        report.volatility_regime = primary_analysis.volatility_regime
        report.current_price = float(primary_df["Close"].iloc[-1])
        report.timestamp = str(primary_df.index[-1])
        report.mtf_signals[self.config.primary_timeframe.value] = primary_analysis.composite_signal
        report.mtf_scores[self.config.primary_timeframe.value] = primary_analysis.composite_score

        # ── 2. Confirmation timeframes ─────────────────────────────
        for tf in self.config.confirmation_timeframes:
            cf_df = dfs.get(tf)
            if cf_df is None or cf_df.empty:
                continue
            try:
                cf_analysis = self.tech_engine.analyze(cf_df, symbol, tf.value)
                report.confirmation_analyses.append(cf_analysis)
                report.mtf_signals[tf.value] = cf_analysis.composite_signal
                report.mtf_scores[tf.value] = cf_analysis.composite_score
            except Exception as e:
                logger.warning(f"Confirmation TF {tf.value} failed for {symbol}: {e}")

        # ── 3. MTF agreement ───────────────────────────────────────
        report.mtf_agreement = self._check_mtf_agreement(report)

        # ── 4. Fundamental context (optional) ──────────────────────
        if use_news:
            news = self._get_news_engine()
            if context is None:
                context = news.get_fundamental_context()
            event_risk, event_desc = news.get_event_risk_level(symbol, context)
            report.event_risk = event_risk
            report.event_description = event_desc
            sentiment, sent_desc = news.get_pair_sentiment(symbol, context)
            report.news_sentiment = sentiment
            report.news_description = sent_desc
            report.news_summary = context.llm_summary
        else:
            report.event_risk = "clear"
            report.event_description = ""
            report.news_sentiment = 0.0
            report.news_description = ""
            report.news_summary = ""

        # ── 5. Spread estimate from primary data (avoids extra API call) ─
        try:
            report.spread_estimate = float(
                (primary_df["High"] - primary_df["Low"]).median()
            )
        except Exception:
            report.spread_estimate = 0.0

        # ── 6. Final signal ────────────────────────────────────────
        self._compute_final_signal(report)

        # ── 6b. Session filter (works for both live and backtest) ─
        if self.config.backtest.session_filter:
            try:
                bar_ts = pd.Timestamp(report.timestamp)
                bar_hour = bar_ts.hour
                bar_wd = bar_ts.weekday()
                _is_xau_pair = report.display_name == "XAU/USD"
                _blocked_hours = (self.config.xau_strategy.blocked_hours_utc
                                  if _is_xau_pair else self.config.backtest.blocked_hours_utc)
                if bar_hour in _blocked_hours:
                    report.final_signal = SignalStrength.NEUTRAL
                    report.signal_quality = "F"
                    report.warnings.append(f"SESSION FILTER: {bar_hour:02}:00 UTC blocked")
                _allowed_wd = (self.config.xau_strategy.allowed_weekdays
                               if _is_xau_pair else self.config.backtest.allowed_weekdays)
                if _allowed_wd and bar_wd not in _allowed_wd:
                    day_names = {0: "Mon", 1: "Tue", 2: "Wed", 3: "Thu",
                                 4: "Fri", 5: "Sat", 6: "Sun"}
                    report.final_signal = SignalStrength.NEUTRAL
                    report.signal_quality = "F"
                    report.warnings.append(
                        f"DAY FILTER: {day_names.get(bar_wd, '?')} not in allowed days"
                    )
            except Exception:
                # Fail-open by design, but never silently (see live wrapper).
                logger.exception(
                    f"{symbol}: session/day filter errored — filter SKIPPED"
                )

        # ── 6c. Transitioning regime quality boost ──────────────────
        if report.market_regime == "transitioning" and report.signal_quality in ("B", "C", "D"):
            abs_score = abs(report.final_score)
            if abs_score >= 30 and report.final_confidence >= 50 and report.mtf_agreement:
                report.signal_quality = "A"
                report.action_recommendation = f"{report.final_signal.name.replace('_', ' ')} — Trend emerging, early entry"
            elif abs_score >= 25 and report.final_confidence >= 45:
                report.signal_quality = "B"
                report.action_recommendation = f"{report.final_signal.name.replace('_', ' ')} — Trend forming, watch closely"

        # ── 6d. Exhausted trend quality cap ───────────────────────
        if (report.market_regime.startswith("trending")
                and report.exhaustion_count >= 2
                and report.signal_quality in ("C", "D")):
            report.warnings.append(
                "EXHAUSTED TREND FILTER: C-grade blocked in exhausted trend"
            )
            report.signal_quality = "D"
            report.trade_setup = None

        # ── 6e. SELL quality gate ─────────────────────────────────
        xau_is_sell_dir = (
            report.final_signal in (SignalStrength.SELL, SignalStrength.STRONG_SELL)
            or (report.final_signal == SignalStrength.NEUTRAL and report.final_score < 0)
        )
        # Actionable grades only: without this guard the gate would "upgrade"
        # a session-blocked F report to D, which re-opens the M5 pullback
        # path (its entry check is quality != "F").
        if (self.config.xau_strategy.enabled
                and report.display_name == "XAU/USD"
                and xau_is_sell_dir
                and report.signal_quality in ("A+", "A", "B", "C")
                and report.signal_quality not in self.config.xau_strategy.sell_min_quality):
            report.warnings.append(
                f"SELL FILTER: quality {report.signal_quality} below minimum for SELL"
            )
            report.signal_quality = "D"
            report.trade_setup = None

        # ── 6f. MTF opposition gate ───────────────────────────────
        # Never trade against a multi-timeframe consensus: if the
        # threshold number of TFs point opposite to the trade direction,
        # block the setup (counter-trend entries in gold bleed).
        if report.signal_quality in ("A+", "A", "B", "C") and report.mtf_signals:
            sigs = list(report.mtf_signals.values())
            bull_n = sum(1 for s in sigs if s in (SignalStrength.BUY, SignalStrength.STRONG_BUY))
            bear_n = sum(1 for s in sigs if s in (SignalStrength.SELL, SignalStrength.STRONG_SELL))
            thr = self.config.signals.mtf_agreement_threshold
            opposed = (bear_n >= thr and report.final_score > 0) or \
                      (bull_n >= thr and report.final_score < 0)
            if opposed:
                report.warnings.append(
                    "MTF OPPOSITION: majority of timeframes point against signal — no trade"
                )
                report.signal_quality = "D"
                report.trade_setup = None

        # ── 6g. SELL volatility gate ──────────────────────────────
        # Shorts in quiet markets fade against gold's macro uptrend;
        # only take sells when volatility regime confirms real selling.
        if (self.config.xau_strategy.enabled
                and report.display_name == "XAU/USD"
                and xau_is_sell_dir
                and report.signal_quality in ("A+", "A", "B", "C")
                and report.volatility_regime
                and report.volatility_regime not in self.config.xau_strategy.sell_vol_regimes):
            report.warnings.append(
                f"SELL FILTER: volatility '{report.volatility_regime}' too quiet for shorts"
            )
            report.signal_quality = "D"
            report.trade_setup = None

        # ── 6g2. XAU regime gates (opt-in via per-TF factory overrides) ──
        if (self.config.xau_strategy.enabled
                and report.display_name == "XAU/USD"
                and report.signal_quality in ("A+", "A", "B", "C")):
            if report.market_regime in self.config.xau_strategy.blocked_regimes:
                report.warnings.append(
                    f"REGIME FILTER: '{report.market_regime}' blocked for XAU"
                )
                report.signal_quality = "D"
                report.trade_setup = None
            elif (xau_is_sell_dir
                    and report.market_regime
                    in self.config.xau_strategy.sell_blocked_regimes):
                report.warnings.append(
                    f"SELL FILTER: regime '{report.market_regime}' blocked for shorts"
                )
                report.signal_quality = "D"
                report.trade_setup = None
            elif xau_is_sell_dir and self.config.xau_strategy.sell_blocked_hours_utc:
                # ── 6g3. XAU SELL hour gate (opt-in via per-TF factories) ──
                # Evaluated on the bar's CLOSE time (= entry time under
                # next-open fills), the same clock the session filter and
                # backtest replay use — NOT the bar label hour.
                try:
                    close_h = (
                        pd.Timestamp(report.timestamp)
                        + pd.Timedelta(
                            seconds=BaseDataProvider.CANDLE_SECONDS[
                                self.config.primary_timeframe
                            ]
                        )
                    ).hour
                except Exception:
                    close_h = None
                    logger.warning(
                        f"{report.symbol}: bar-close hour underivable — "
                        f"SELL hour gate FAILS CLOSED (short blocked)"
                    )
                # Fail CLOSED: an underivable close hour blocks the short
                # (`None in [15, 17]` used to pass silently). Never fires
                # on parseable timestamps, so pinned books are unchanged.
                if (close_h is None
                        or close_h in self.config.xau_strategy.sell_blocked_hours_utc):
                    hour_txt = "??" if close_h is None else f"{close_h:02}"
                    report.warnings.append(
                        f"SELL FILTER: {hour_txt}:00 UTC entry blocked for shorts"
                    )
                    report.signal_quality = "D"
                    report.trade_setup = None

        # ── 6h. Big-candle entry gate (opt-in via per-TF factories) ──
        # An abnormally large candle in the recent window = institutional
        # burst; price may reverse violently, so NO entry of any kind may
        # fire. Detection runs for ANY grade: the warning is the
        # suppression channel for the XAU_PB path (which accepts D) and
        # the XAU_CP path (which accepts even F). The D-downgrade itself
        # touches actionable grades only — F stays F (same convention as
        # 6e).
        if (self.config.xau_strategy.enabled
                and report.display_name == "XAU/USD"
                and self.config.xau_strategy.big_candle_atr_multiple > 0):
            big_candle = self._detect_big_candle(primary_df)
            if big_candle is not None:
                bc_ts, bc_ratio = big_candle
                report.warnings.append(
                    f"BIG CANDLE FILTER: {bc_ratio:.1f}x ATR bar at {bc_ts} "
                    f"within "
                    f"{self.config.xau_strategy.big_candle_lookback_hours:g}h "
                    f"— no entries"
                )
                if report.signal_quality in ("A+", "A", "B", "C"):
                    report.signal_quality = "D"
                    report.trade_setup = None

        # ── 6i. M5 noise gate (gold-tuned thresholds — XAU only) ──
        if (self.config.primary_timeframe == Timeframe.M5
                and report.display_name == "XAU/USD"):
            self._apply_m5_filters(report)

        # ── 7. Trade setup for actionable signals ──────────────────
        self._attach_core_setup(report, pip_value)

        # ── 8. Reasons ─────────────────────────────────────────────
        self._collect_reasons(report)

        # ── 9. M5 pullback entry (XAU_PB) ──────────────────────────
        # Second archetype: when no breakout setup fired, look for a
        # trend-continuation pullback (retrace to EMA21/KC-mid). Runs
        # after reasons so its marker lands at reasons[0] for tagging.
        if (self.config.primary_timeframe == Timeframe.M5
                and report.display_name == "XAU/USD"
                and self.config.xau_strategy.m5_pullback_enabled
                and report.trade_setup is None
                and report.signal_quality != "F"
                and not any(w.startswith("BIG CANDLE FILTER")
                            for w in report.warnings)):
            self._try_m5_pullback(report, primary_df, pip_value)

        # ── 10. Chart-pattern entry (XAU_CP) ────────────────────────
        # Third archetype: completion of a classical chart pattern
        # (double top/bottom, H&S, flag — see chart_patterns.py). Only
        # fills empty slots (core + pullback setups take precedence).
        # Unlike XAU_PB this must NOT require quality != "F": a weak
        # composite (grade F on |score| < 15) is the normal state at a
        # reversal-pattern completion. Session/day blocks still bind —
        # they are detected by their warnings (and the live wrapper +
        # backtest replay apply the bar-close session predicate again
        # on top, so a pattern setup can never trade a blocked hour).
        if (report.display_name == "XAU/USD"
                and self.config.xau_strategy.cp_enabled
                and report.trade_setup is None
                and not any(w.startswith(("SESSION FILTER", "DAY FILTER",
                                          "WEEKEND FILTER",
                                          "BIG CANDLE FILTER"))
                            for w in report.warnings)):
            self._try_chart_pattern(report, primary_df, pip_value)

        # ── 11. Stale spread-gate vetoes ─────────────────────────────
        # Whichever archetype shipped the setup, a core generator's
        # cost-floor veto no longer describes it.
        self._drop_stale_spread_gate_vetoes(report)

        return report

    # ── Internals ──────────────────────────────────────────────────

    # Only A+/A reach the XAU ATR generator (gate at the call site).
    _TP_QUALITY_SCALE = {"A+": 2.0, "A": 1.5}

    def _attach_core_setup(self, report: SignalReport, pip_value: float) -> None:
        """Step 7: the XAU ATR generator (A+/A) first, the swing generator
        (GENERIC) for anything it declined. A veto either one leaves
        behind is reconciled after every archetype has run
        (`_drop_stale_spread_gate_vetoes`)."""
        # ATR generator is XAU-only: it hardcodes gold pip value and
        # 2-decimal price rounding.
        if (report.trade_setup is None and report.signal_quality in ("A+", "A")
                and report.display_name == "XAU/USD"):
            atr = (report.primary_analysis.key_levels.get("atr", 0)
                   if report.primary_analysis else 0)
            if atr > 0:
                report.trade_setup = self._generate_xau_atr_trade_setup(report, atr)
        if report.trade_setup is None and report.signal_quality in ("A+", "A", "B", "C"):
            report.trade_setup = self._generate_trade_setup(report, pip_value)
        # A shipped setup always names a direction. The signal thresholds
        # (BUY/SELL need confidence >= 50) sit above the C grade floor
        # (>= 45), and gate 6c grants B at >= 45 in "transitioning", so a
        # C/B report could leave here as NEUTRAL with a full setup: the
        # backtester traded it (the replay never reads final_signal) while
        # the Telegram alert gate rejects NEUTRAL first (2026-09-13 audit:
        # 5/91 H1 12-mo book trades, +3.7R of +22.2R, were never alerted).
        # The generators derived the direction from the score sign — the
        # same sign every direction gate above used — so this changes
        # display and alert eligibility only; the trade books are unchanged.
        # A report no generator served stays NEUTRAL, as before.
        if (report.trade_setup is not None
                and report.final_signal == SignalStrength.NEUTRAL):
            report.final_signal = (SignalStrength.BUY
                                   if report.trade_setup.direction == "BUY"
                                   else SignalStrength.SELL)
            # The action text was written before the gate chain with the
            # old signal name; keep it consistent with the header.
            if report.action_recommendation.startswith("NEUTRAL"):
                report.action_recommendation = (
                    report.final_signal.name.replace("_", " ")
                    + report.action_recommendation[len("NEUTRAL"):])

    @staticmethod
    def _drop_stale_spread_gate_vetoes(report: SignalReport) -> None:
        """Step 11: runs once after every setup archetype (core ATR/swing,
        XAU_PB, XAU_CP) has had its turn.

        Both core generators veto a target under the cost floor with a
        `_SPREAD_GATE_PREFIX` warning. When a later generator then ships
        a setup with a bigger target, that veto is stale — the 2026-09-08
        07:45 UTC A-grade alert said "costs would eat the edge" beside a
        $26.49 TP1. A stale veto moves to `cleaned_warnings` (metadata
        only, the XAU_CP cleanup's contract; that path has already
        stripped its own copy by the time this runs, and the PB path is
        covered here). Nothing downstream reads warnings while a setup
        exists, so trade books are unchanged — verified 2026-09-08 on the
        3-yr H1/M15 books; only the exported `warnings` text column
        differs.
        """
        if report.trade_setup is None:
            return
        kept: List[str] = []
        stale: List[str] = []
        for warning in report.warnings:
            (stale if warning.startswith(_SPREAD_GATE_PREFIX) else kept).append(warning)
        if stale:
            report.warnings[:] = kept
            report.cleaned_warnings.extend(stale)

    def _generate_xau_atr_trade_setup(
        self, report: SignalReport, atr: float
    ) -> Optional[TradeSetup]:
        """XAU ATR-based trade setup with quality-tiered TP scaling."""
        cfg = self.config.xau_strategy
        price = report.current_price
        pip_value = 0.10

        if report.final_signal in (SignalStrength.BUY, SignalStrength.STRONG_BUY):
            is_buy = True
        elif report.final_signal in (SignalStrength.SELL, SignalStrength.STRONG_SELL):
            is_buy = False
        else:
            is_buy = report.final_score >= 0

        quality_scale = self._TP_QUALITY_SCALE[report.signal_quality]
        if report.signal_quality == "A" and report.exhaustion_count >= 2:
            quality_scale = 1.0
        tp_mult = cfg.tp_atr_multiplier * quality_scale

        if is_buy:
            sl = price - cfg.sl_atr_multiplier * atr
            tp1 = price + tp_mult * atr
            tp2 = price + tp_mult * 1.5 * atr
            direction = "BUY"
        else:
            sl = price + cfg.sl_atr_multiplier * atr
            tp1 = price - tp_mult * atr
            tp2 = price - tp_mult * 1.5 * atr
            direction = "SELL"

        tf = self.config.signals.tp_tightening_factor
        tp1 = price + (tp1 - price) * tf
        tp2 = price + (tp2 - price) * tf

        sl_distance = abs(price - sl)
        tp_distance = abs(tp1 - price)
        rr = tp_distance / sl_distance if sl_distance > 0 else 0

        min_rr = self.config.signals.min_rr_ratio
        if rr < min_rr - 1e-9:
            return None

        # Spread gate (same rule as _generate_trade_setup).
        min_tp = self.config.signals.round_trip_cost * self.config.signals.min_tp_cost_multiple
        if tp_distance < min_tp:
            report.warnings.append(
                f"{_SPREAD_GATE_PREFIX} target ${tp_distance:.2f} < ${min_tp:.2f} minimum — "
                f"costs would eat the edge"
            )
            return None

        return TradeSetup(
            pair=report.symbol,
            display_name=report.display_name,
            direction=direction,
            entry_price=price,
            stop_loss=round(sl, 2),
            take_profit_1=round(tp1, 2),
            take_profit_2=round(tp2, 2),
            risk_reward=round(rr, 2),
            risk_pips=round(sl_distance / pip_value, 1),
            reward_pips=round(tp_distance / pip_value, 1),
            pip_value=pip_value,
            strategy="XAU_ATR",
            round_trip_cost=self.config.signals.round_trip_cost,
            trail_arm_r=self.config.signals.trail_arm_r,
            trail_dist_r=self.config.signals.trail_dist_r,
        )

    def _check_mtf_agreement(self, report: SignalReport) -> bool:
        signals = list(report.mtf_signals.values())
        if len(signals) < 2:
            return False

        bullish = sum(1 for s in signals if s in (SignalStrength.BUY, SignalStrength.STRONG_BUY))
        bearish = sum(1 for s in signals if s in (SignalStrength.SELL, SignalStrength.STRONG_SELL))

        threshold = self.config.signals.mtf_agreement_threshold
        # Agreement must point the same way as the primary timeframe —
        # two bullish confirmation TFs must not "confirm" a bearish primary.
        if report.primary_score > 0:
            return bullish >= threshold
        if report.primary_score < 0:
            return bearish >= threshold
        return bullish >= threshold or bearish >= threshold

    def _compute_final_signal(self, report: SignalReport):
        score = report.primary_score
        confidence = report.primary_confidence

        # ── Bonuses (additive, applied first) ─────────────────────
        if report.mtf_agreement:
            confidence += 10
            all_scores = list(report.mtf_scores.values())
            if all_scores:
                mtf_avg = sum(all_scores) / len(all_scores)
                score = score * 0.6 + mtf_avg * 0.4

        if self.config.news.sentiment_affects_signals and abs(report.news_sentiment) > 0.2:
            if (report.news_sentiment > 0 and score > 0) or \
               (report.news_sentiment < 0 and score < 0):
                confidence += 5

        primary = report.primary_analysis
        if primary and report.market_regime.startswith("trending"):
            rsi_ind = primary.indicators.get("rsi")
            bb_ind = primary.indicators.get("bollinger")
            # Pullback bonuses require a KNOWN trend direction — the mixed
            # "trending" regime (EMAs unaligned) gets no directional bonus.
            if rsi_ind:
                rsi_val = rsi_ind.value
                if report.market_regime == "trending_up" and score > 0 and 38 <= rsi_val <= 55:
                    confidence += 5
                elif report.market_regime == "trending_down" and score < 0 and 45 <= rsi_val <= 62:
                    confidence += 5
            if bb_ind:
                bb_pos = bb_ind.value
                if report.market_regime == "trending_up" and score > 0 and bb_pos < 0.4:
                    confidence += 5
                elif report.market_regime == "trending_down" and score < 0 and bb_pos > 0.6:
                    confidence += 5

        # ── Event risk (hard cap, separate from penalty system) ───
        # Warning always shown; the confidence cap is config-gated
        # (news.event_risk_caps_signals) — see NewsConfig for the audit
        # showing the cap costs R on XAU.
        if report.event_risk == "danger":
            report.warnings.append(f"EVENT RISK: {report.event_description}")
            if self.config.news.event_risk_caps_signals:
                confidence = min(confidence, 30)
                report.action_recommendation = f"WAIT — {report.event_description}"

        # ── Penalties (diminishing cap system) ────────────────────
        # Collect all applicable penalties, then apply with exponential
        # decay so stacking never guts a signal beyond 30%.
        penalties = []

        if not report.mtf_agreement:
            penalties.append(("mtf_disagreement", 15))
            if len(report.mtf_scores) >= 2:
                scores_list = list(report.mtf_scores.values())
                if any(s > 10 for s in scores_list) and any(s < -10 for s in scores_list):
                    report.warnings.append("Timeframes CONFLICT — mixed signals across timeframes")

        if abs(report.news_sentiment) > 0.2:
            if (report.news_sentiment > 0 and score < -10) or \
               (report.news_sentiment < 0 and score > 10):
                report.warnings.append("News sentiment CONFLICTS with technical signal")
                if self.config.news.sentiment_affects_signals:
                    penalties.append(("news_conflict", 10))

        if report.event_risk == "caution":
            report.warnings.append(f"Upcoming event: {report.event_description}")
            if self.config.news.event_risk_caps_signals:
                penalties.append(("event_caution", 15))

        is_counter_trend = (
            (report.market_regime == "trending_up" and score < 0) or
            (report.market_regime == "trending_down" and score > 0)
        )
        if is_counter_trend:
            report.warnings.append("Counter-trend signal — reduced confidence")
            penalties.append(("counter_trend", 15))

        if report.volatility_regime == "extreme":
            report.warnings.append("EXTREME volatility — wider stops needed")
            penalties.append(("extreme_vol", 10))
        elif report.volatility_regime == "low":
            report.warnings.append("Low volatility — potential for breakout")

        if primary is not None:
            ema200 = primary.key_levels.get("ema_trend")
            if ema200 and ema200 > 0 and report.current_price > 0:
                ema_dist_pct = abs(report.current_price - ema200) / ema200 * 100.0
                # FB2_EMA_LO: research override for the EMA200-proximity
                # penalty band; the tuned default is 0.5. An unparseable
                # value must degrade to the default, never crash the scan
                # (2026-08-04 audit; hardened per owner delegation).
                try:
                    _lo = float(os.environ.get("FB2_EMA_LO", "0.5"))
                except (TypeError, ValueError):
                    logger.warning(
                        "FB2_EMA_LO=%r is not a number — using tuned "
                        "default 0.5", os.environ.get("FB2_EMA_LO"))
                    _lo = 0.5
                if ema_dist_pct < _lo:
                    report.warnings.append(f"NEAR EMA200 ({ema_dist_pct:.2f}%) — chop risk")
                    penalties.append(("ema200_near", 5))
                elif ema_dist_pct > 2.5:
                    report.warnings.append(f"FAR FROM EMA200 ({ema_dist_pct:.2f}%) — over-extended")
                    penalties.append(("ema200_far", 5))

            atr = primary.key_levels.get("atr", 0)
            if atr > 0 and score != 0:
                # Distance is measured from the FARTHEST of the 3 tracked
                # swings (sorted()[0] = deepest low / highest high), so the
                # penalty fires earlier than a nearest-swing rule would.
                # Calibrated as-is in every validated book — changing the
                # anchor is a signal-logic change requiring re-validation.
                if score > 0:
                    swing_lows = sorted(
                        v for k, v in primary.key_levels.items() if k.startswith("swing_low_")
                    )
                    if swing_lows and report.current_price - swing_lows[0] > 3.5 * atr:
                        report.warnings.append("TREND EXHAUSTION — price extended >3.5 ATR past swing low")
                        penalties.append(("exhaustion", 8))
                else:
                    swing_highs = sorted(
                        (v for k, v in primary.key_levels.items() if k.startswith("swing_high_")),
                        reverse=True,
                    )
                    if swing_highs and swing_highs[0] - report.current_price > 3.5 * atr:
                        report.warnings.append("TREND EXHAUSTION — price extended >3.5 ATR past swing high")
                        penalties.append(("exhaustion", 8))

            ema_medium = primary.key_levels.get("ema_medium")
            if atr > 0 and ema_medium and report.market_regime.startswith("trending"):
                pullback_dist = abs(report.current_price - ema_medium) / atr
                if pullback_dist > 2.5:
                    report.warnings.append(f"NO PULLBACK — price {pullback_dist:.1f} ATR from EMA21")
                    penalties.append(("no_pullback", 8))
                elif pullback_dist > 2.0:
                    report.warnings.append(f"Weak pullback — price {pullback_dist:.1f} ATR from EMA21")
                    penalties.append(("weak_pullback", 5))

            if report.market_regime.startswith("trending"):
                rsi_ind = primary.indicators.get("rsi")
                bb_ind = primary.indicators.get("bollinger")
                # Extension penalties apply in ANY strong-ADX regime; the
                # mixed "trending" regime penalizes BOTH directions (it
                # previously penalized only BUYs — the SELL side of the
                # same bullish-bias bug fixed in the RSI reinterpretation).
                if rsi_ind:
                    rsi_val = rsi_ind.value
                    if report.market_regime in ("trending_up", "trending") and score > 0 and rsi_val > 65:
                        report.warnings.append(f"Extended — RSI {rsi_val:.0f} in uptrend")
                        penalties.append(("rsi_extended", 5))
                    elif report.market_regime in ("trending_down", "trending") and score < 0 and rsi_val < 35:
                        report.warnings.append(f"Extended — RSI {rsi_val:.0f} in downtrend")
                        penalties.append(("rsi_extended", 5))
                if bb_ind:
                    bb_pos = bb_ind.value
                    if report.market_regime in ("trending_up", "trending") and score > 0 and bb_pos > 0.85:
                        penalties.append(("bb_overextended", 8))
                    elif report.market_regime in ("trending_down", "trending") and score < 0 and bb_pos < 0.15:
                        penalties.append(("bb_overextended", 8))

        # Hard block: static exhaustion + fading momentum → kill the trade
        # Static signals alone are normal in gold trends. Only block when
        # price is extended AND momentum indicators confirm deceleration.
        _EXHAUSTION_TAGS = {"exhaustion", "no_pullback", "rsi_extended", "bb_overextended"}
        exhaustion_count = sum(1 for name, _ in penalties if name in _EXHAUSTION_TAGS)
        report.exhaustion_count = exhaustion_count
        momentum_fading = False
        if primary is not None and exhaustion_count >= 2:
            roc_ind = primary.indicators.get("roc")
            macd_ind = primary.indicators.get("macd")
            roc_fading = False
            macd_fading = False
            if roc_ind:
                if score > 0 and roc_ind.value < 0:
                    roc_fading = True
                elif score < 0 and roc_ind.value > 0:
                    roc_fading = True
            if macd_ind:
                if score > 0 and macd_ind.signal in (SignalStrength.SELL, SignalStrength.STRONG_SELL):
                    macd_fading = True
                elif score < 0 and macd_ind.signal in (SignalStrength.BUY, SignalStrength.STRONG_BUY):
                    macd_fading = True
            momentum_fading = roc_fading or macd_fading
        exhaustion_blocked = exhaustion_count >= 2 and momentum_fading

        MAX_PENALTY = 30.0
        penalties.sort(key=lambda x: x[1], reverse=True)
        total_penalty = 0.0
        decay = 1.0
        for _name, amount in penalties:
            applied = amount * decay
            if total_penalty + applied > MAX_PENALTY:
                applied = MAX_PENALTY - total_penalty
            total_penalty += applied
            decay *= 0.5
            if total_penalty >= MAX_PENALTY:
                break

        # Record-keeping for position_sizing / signal-log forensics.
        # Purely additive — nothing below reads these fields.
        report.penalty_tags = {name: float(amount) for name, amount in penalties}
        report.penalty_total = float(total_penalty)
        report.momentum_fading = bool(momentum_fading)

        confidence -= total_penalty
        confidence = max(0.0, min(95.0, confidence))

        if exhaustion_blocked:
            report.warnings.append(
                f"EXHAUSTION BLOCK — {exhaustion_count} exhaustion signals fired, trade rejected"
            )
            confidence = 0.0

        # Final signal
        if score >= 35 and confidence >= 55:
            report.final_signal = SignalStrength.STRONG_BUY
        elif score >= 25 and confidence >= 50:
            report.final_signal = SignalStrength.BUY
        elif score <= -35 and confidence >= 55:
            report.final_signal = SignalStrength.STRONG_SELL
        elif score <= -25 and confidence >= 50:
            report.final_signal = SignalStrength.SELL
        else:
            report.final_signal = SignalStrength.NEUTRAL

        report.final_score = score
        report.final_confidence = confidence

        # Quality grade
        abs_score = abs(score)
        if abs_score >= 40 and confidence >= 65 and report.mtf_agreement:
            report.signal_quality = "A+"
        elif abs_score >= 33 and confidence >= 55 and report.mtf_agreement:
            report.signal_quality = "A"
        elif abs_score >= 28 and confidence >= 50:
            report.signal_quality = "B"
        elif abs_score >= 25 and confidence >= 45:
            report.signal_quality = "C"
        elif abs_score >= 15:
            report.signal_quality = "D"
        else:
            report.signal_quality = "F"

        if not report.action_recommendation:
            sig_name = report.final_signal.name.replace("_", " ")
            if report.signal_quality in ("A+", "A"):
                report.action_recommendation = f"{sig_name} — High-quality setup, consider entry"
            elif report.signal_quality == "B":
                report.action_recommendation = f"{sig_name} — Decent setup, wait for confirmation"
            elif report.signal_quality in ("C", "D"):
                report.action_recommendation = "WATCH — Signal present but low confidence"
            else:
                report.action_recommendation = "NO TRADE — No clear signal"

    def _generate_trade_setup(self, report: SignalReport, pip_value: float) -> Optional[TradeSetup]:
        if not report.primary_analysis:
            return None

        price = report.current_price
        atr = report.primary_analysis.key_levels.get("atr", 0)
        if atr == 0:
            return None
        if pip_value <= 0:
            return None

        # Fix 2: Minimum ATR threshold for XAU/USD
        if report.display_name == "XAU/USD" and atr < self.config.xau_strategy.min_atr_threshold:
            report.warnings.append(
                f"LOW ATR FILTER: ATR ${atr:.2f} < ${self.config.xau_strategy.min_atr_threshold:.2f} minimum"
            )
            return None

        if report.final_signal in (SignalStrength.BUY, SignalStrength.STRONG_BUY):
            is_buy = True
        elif report.final_signal in (SignalStrength.SELL, SignalStrength.STRONG_SELL):
            is_buy = False
        else:
            is_buy = report.final_score >= 0
        levels = report.primary_analysis.key_levels

        min_sl_price = self.config.signals.min_sl_pips * pip_value
        min_rr = self.config.signals.min_rr_ratio
        tp_floor = max(min_rr, self.config.signals.tp_floor_rr)
        tf = self.config.signals.tp_tightening_factor
        pre_tight_min_rr = tp_floor / tf if tf > 0 else tp_floor

        tp_mult = 1.5
        tp_swing_cap = 2.5

        sl_atr_floor = (self.config.xau_strategy.v2_sl_atr_multiplier
                        if report.display_name == "XAU/USD" else 1.5)
        sl_swing_buffer = 1.0

        if is_buy:
            swing_lows = sorted(
                [v for k, v in levels.items() if k.startswith("swing_low_") and v < price],
                reverse=True,
            )
            if swing_lows:
                sl_price = swing_lows[0] - atr * sl_swing_buffer
            else:
                sl_price = price - atr * (1.0 + sl_swing_buffer)
            min_sl = max(atr * sl_atr_floor, min_sl_price)
            if price - sl_price < min_sl:
                sl_price = price - min_sl

            sl_distance = price - sl_price

            swing_highs = sorted(
                v for k, v in levels.items() if k.startswith("swing_high_") and v > price + atr * 0.3
            )
            if (swing_highs
                    and swing_highs[0] - price <= atr * tp_swing_cap
                    and swing_highs[0] - price >= sl_distance * pre_tight_min_rr):
                tp1 = swing_highs[0]
            else:
                tp1 = price + max(atr * tp_mult, sl_distance * pre_tight_min_rr)
            if len(swing_highs) >= 2 and swing_highs[1] - price <= atr * 5.0:
                tp2 = swing_highs[1]
            else:
                tp2 = price + max(atr * (tp_mult + 1.5), sl_distance * pre_tight_min_rr * 1.5)

            direction = "BUY"
            entry, sl = price, sl_price
        else:
            swing_highs = sorted(
                v for k, v in levels.items() if k.startswith("swing_high_") and v > price
            )
            if swing_highs:
                sl_price = swing_highs[0] + atr * sl_swing_buffer
            else:
                sl_price = price + atr * (1.0 + sl_swing_buffer)
            min_sl = max(atr * sl_atr_floor, min_sl_price)
            if sl_price - price < min_sl:
                sl_price = price + min_sl

            sl_distance = sl_price - price

            swing_lows = sorted(
                [v for k, v in levels.items() if k.startswith("swing_low_") and v < price - atr * 0.3],
                reverse=True,
            )
            if (swing_lows
                    and price - swing_lows[0] <= atr * tp_swing_cap
                    and price - swing_lows[0] >= sl_distance * pre_tight_min_rr):
                tp1 = swing_lows[0]
            else:
                tp1 = price - max(atr * tp_mult, sl_distance * pre_tight_min_rr)
            if len(swing_lows) >= 2 and price - swing_lows[1] <= atr * 5.0:
                tp2 = swing_lows[1]
            else:
                tp2 = price - max(atr * (tp_mult + 1.5), sl_distance * pre_tight_min_rr * 1.5)

            direction = "SELL"
            entry, sl = price, sl_price

        tf = self.config.signals.tp_tightening_factor
        tp1 = entry + (tp1 - entry) * tf
        tp2 = entry + (tp2 - entry) * tf

        # TP2 must sit beyond TP1 — swing picks can invert when TP1 falls
        # back to the ATR target while TP2 takes a nearer second swing.
        if is_buy:
            tp2 = max(tp2, tp1 + atr)
        else:
            tp2 = min(tp2, tp1 - atr)

        tp1_distance = abs(tp1 - entry)
        rr = tp1_distance / sl_distance if sl_distance > 0 else 0

        if rr < min_rr - 1e-9:
            return None

        # Spread gate: a target too close to entry hands most of the win
        # to transaction costs (validated on M5: sub-$10 targets lose).
        min_tp = self.config.signals.round_trip_cost * self.config.signals.min_tp_cost_multiple
        if tp1_distance < min_tp:
            report.warnings.append(
                f"{_SPREAD_GATE_PREFIX} target ${tp1_distance:.2f} < ${min_tp:.2f} minimum — "
                f"costs would eat the edge"
            )
            return None

        risk_pips = sl_distance / pip_value
        reward_pips = tp1_distance / pip_value

        return TradeSetup(
            pair=report.symbol,
            display_name=report.display_name,
            direction=direction,
            entry_price=entry,
            stop_loss=sl,
            take_profit_1=tp1,
            take_profit_2=tp2,
            risk_reward=round(rr, 2),
            risk_pips=risk_pips,
            reward_pips=reward_pips,
            pip_value=pip_value,
            round_trip_cost=self.config.signals.round_trip_cost,
            trail_arm_r=self.config.signals.trail_arm_r,
            trail_dist_r=self.config.signals.trail_dist_r,
        )

    def _collect_reasons(self, report: SignalReport):
        if not report.primary_analysis:
            return

        for name, ind in report.primary_analysis.indicators.items():
            if name.startswith("_") or ind.weight == 0:
                continue
            if ind.signal in (SignalStrength.BUY, SignalStrength.STRONG_BUY):
                prefix = "STRONG: " if ind.signal == SignalStrength.STRONG_BUY else ""
                report.bullish_reasons.append(f"{prefix}{ind.name}: {ind.description}")
            elif ind.signal in (SignalStrength.SELL, SignalStrength.STRONG_SELL):
                prefix = "STRONG: " if ind.signal == SignalStrength.STRONG_SELL else ""
                report.bearish_reasons.append(f"{prefix}{ind.name}: {ind.description}")

        if report.news_sentiment > 0.1:
            report.bullish_reasons.append(f"News: {report.news_description}")
        elif report.news_sentiment < -0.1:
            report.bearish_reasons.append(f"News: {report.news_description}")