main.py

import os, sys, traceback, pandas as pd
from main_phase4 import load_and_clean_slate, score_player_ace_mode, optimize_single_lineup
from advanced_optimizer import AdvancedOptimizer
from portfolio_manager import PortfolioManager
from output_formatter import OutputFormatter

# FanDuel NBA Classic display order
SLOT_NAMES = ["PG", "PG", "SG", "SG", "SF", "SF", "PF", "PF", "C"]


def validate_lineup(lp: pd.DataFrame):
    errors = []
    if len(lp) != 9:
        errors.append(f"Wrong size: {len(lp)} players (need 9)")
    if lp["Id"].duplicated().any():
        errors.append(f"Duplicate IDs: {lp[lp['Id'].duplicated()]['Name'].tolist()}")
    if lp["Salary"].sum() > 60000:
        errors.append(f"Over cap: ${lp['Salary'].sum():,.0f}")
    return len(errors) == 0, errors


def print_lineup(index: int, lp: pd.DataFrame):
    """Print one lineup in FanDuel display order with matchup intel."""
    ordered = lp.sort_values("_assigned_slot_idx").reset_index(drop=True)

    total_sal  = ordered["Salary"].sum()
    total_fppg = ordered["FPPG"].sum()
    total_ace  = ordered["AceScore"].sum()

    print(f"\n{'─'*72}")
    print(f"  Lineup {index:>2}  |  Salary: ${total_sal:>6,.0f}  |  "
          f"FPPG: {total_fppg:>6.1f}  |  AceScore: {total_ace:>7.2f}")
    print(f"{'─'*72}")
    print(f"  {'Slot':<5} {'Name':<24} {'Pos':<8} {'Salary':>7}  "
          f"{'FPPG':>6}  {'Ace':>7}  {'Matchup'}")
    print(f"  {'─'*5} {'─'*24} {'─'*8} {'─'*7}  {'─'*6}  {'─'*7}  {'─'*12}")

    for i, row in ordered.iterrows():
        slot       = SLOT_NAMES[int(row.get("_assigned_slot_idx", i))] if i < len(SLOT_NAMES) else "?"
        name       = str(row.get("Name", ""))[:24]
        pos        = str(row.get("Position", ""))[:8]
        salary     = int(row.get("Salary", 0))
        fppg       = float(row.get("FPPG", 0))
        ace        = float(row.get("AceScore", 0))
        grade      = row.get("MatchupGrade", 5.0)
        label      = str(row.get("MatchupLabel", "") or "")
        opp        = str(row.get("Opponent", "") or "")

        matchup_str = f"vs {opp:<4} {grade:.1f}/10 {label}"

        print(f"  {slot:<5} {name:<24} {pos:<8} ${salary:>6,}  "
              f"{fppg:>6.1f}  {ace:>7.2f}  {matchup_str}")

    print(f"  {'─'*5} {'─'*24} {'─'*8} {'─'*7}  {'─'*6}  {'─'*7}")
    print(f"  {'TOTAL':<5} {'':<24} {'':<8} ${total_sal:>6,}  "
          f"{total_fppg:>6.1f}  {total_ace:>7.2f}")


def main():
    try:
        print("=" * 72)
        print("  SPORTY THIEVES — Ace Mode")
        print("=" * 72)

        # ── 1. Load + Clean Slate ─────────────────────────────────────────────
        slate_df = load_and_clean_slate()

        # ── 2. Leverage / Ownership Model ────────────────────────────────────
        adv = AdvancedOptimizer(slate_df)
        slate_df = adv.calculate_ownership_and_leverage()

        # ── 3. AceScore (now includes matchup bonus) ──────────────────────────
        if "AceScore" not in slate_df.columns:
            slate_df["AceScore"] = slate_df.apply(
                lambda r: score_player_ace_mode(r, slate_df), axis=1
            )

        # Print top-10 AceScore players for reference
        top10 = slate_df.nlargest(10, "AceScore")[
            ["Name", "Position", "Opponent", "Salary", "FPPG", "AceScore",
             "MatchupGrade", "MatchupLabel"]
        ]
        print("\n── Top 10 AceScore Players ──────────────────────────────────────────")
        print(f"  {'Name':<24} {'Pos':<8} {'Opp':<5} {'Salary':>7}  "
              f"{'FPPG':>6}  {'Ace':>7}  {'Matchup'}")
        print(f"  {'─'*24} {'─'*8} {'─'*5} {'─'*7}  {'─'*6}  {'─'*7}  {'─'*12}")
        for _, r in top10.iterrows():
            print(f"  {str(r['Name']):<24} {str(r['Position']):<8} "
                  f"{str(r['Opponent']):<5} ${int(r['Salary']):>6,}  "
                  f"{r['FPPG']:>6.1f}  {r['AceScore']:>7.2f}  "
                  f"vs {r['Opponent']} {r['MatchupGrade']:.1f}/10 {r.get('MatchupLabel','')}")

        # ── 4. Build Portfolio ────────────────────────────────────────────────
        print("\n── Building 20-Lineup Portfolio ─────────────────────────────────────")
        pm          = PortfolioManager(slate_df, num_lineups=20)
        raw_portfolio = pm.build_portfolio(None, slate_df)

        valid_portfolio = []
        for lp in raw_portfolio:
            is_valid, errs = validate_lineup(lp)
            if is_valid:
                valid_portfolio.append(lp)
            else:
                print(f"  [SKIP] Lineup rejected: {errs}")

        # Emergency bridge
        if not valid_portfolio:
            print("\n  ⚠️  All lineups rejected — attempting emergency build...")
            emergency = optimize_single_lineup(slate_df, floor=50000)
            if not emergency.empty:
                valid_portfolio = [emergency]

        if not valid_portfolio:
            print("CRITICAL: Even emergency build failed. Check player pool size.")
            return

        # ── 5. Print Lineups ──────────────────────────────────────────────────
        print(f"\n── {len(valid_portfolio)} Valid Lineups ───────────────────────────────────────────")
        for i, lp in enumerate(valid_portfolio, 1):
            print_lineup(i, lp)

        # ── 6. Export ─────────────────────────────────────────────────────────
        formatter  = OutputFormatter()
        lineups_df = formatter.format_for_fanduel_upload(valid_portfolio)
        lineups_df.to_csv("fd_lineups_upload.csv", index=False)
        print(f"\n✅ Exported {len(valid_portfolio)} lineups to fd_lineups_upload.csv")

        # Exposure report
        exposure_df = formatter.calculate_exposures(valid_portfolio)
        exposure_df.to_csv("lineup_exposures.csv", index=False)
        print(f"✅ Exposure report saved to lineup_exposures.csv")

        print("\n── Top Exposure ──────────────────────────────────────────────────────")
        print(exposure_df.head(15).to_string(index=False))

    except SystemExit as e:
        print(f"\n[FATAL] {e}")
    except Exception:
        traceback.print_exc()


if __name__ == "__main__":
    main()