"""Derive lean_mass and fat_mass from weight_log + body_fat_pct for caliper entries."""
from datetime import date, datetime
from decimal import Decimal
from typing import Any, Dict, List, Optional, Tuple

from db import get_cursor, r2d


def as_date(d: Any) -> Optional[date]:
    if d is None:
        return None
    if isinstance(d, datetime):
        return d.date()
    if isinstance(d, date):
        return d
    if isinstance(d, str) and len(d) >= 10:
        return datetime.strptime(d[:10], "%Y-%m-%d").date()
    return None


def _floaty(x: Any) -> float:
    if isinstance(x, Decimal):
        return float(x)
    return float(x)


def load_weight_rows(conn: Any, profile_id: str) -> List[Dict[str, Any]]:
    cur = get_cursor(conn)
    cur.execute(
        "SELECT date, weight FROM weight_log WHERE profile_id=%s ORDER BY date",
        (profile_id,),
    )
    return [r2d(r) for r in cur.fetchall()]


def nearest_weight_kg(weight_rows: List[Dict[str, Any]], caliper_date: Any) -> Optional[float]:
    td = as_date(caliper_date)
    if td is None or not weight_rows:
        return None
    best: Optional[float] = None
    best_delta: Optional[int] = None
    for row in weight_rows:
        wd = as_date(row.get("date"))
        if wd is None:
            continue
        wraw = row.get("weight")
        if wraw is None:
            continue
        delta = abs((wd - td).days)
        if best_delta is None or delta < best_delta:
            best_delta = delta
            best = _floaty(wraw)
    return best


def nearest_weight_kg_from_map(wlog: Dict[Any, Any], target_date: Any) -> Optional[float]:
    if not wlog:
        return None
    td = as_date(target_date)
    if td is None:
        return None
    best_w: Optional[float] = None
    best_delta: Optional[int] = None
    for wd, wval in wlog.items():
        if wval is None:
            continue
        wdate = as_date(wd)
        if wdate is None:
            continue
        delta = abs((wdate - td).days)
        if best_delta is None or delta < best_delta:
            best_delta = delta
            best_w = _floaty(wval)
    return best_w


def compute_lean_fat_kg(weight_kg: float, body_fat_pct: float) -> Tuple[float, float]:
    lean = round(weight_kg - (weight_kg * body_fat_pct / 100.0), 2)
    fat = round(weight_kg * body_fat_pct / 100.0, 2)
    return lean, fat


def fill_caliper_body_comp(d: Dict[str, Any], conn: Any, profile_id: str) -> None:
    """Mutate caliper API payload (model_dump) before INSERT/UPDATE."""
    bf = d.get("body_fat_pct")
    if bf is None:
        return
    if d.get("lean_mass") is not None and d.get("fat_mass") is not None:
        return
    rows = load_weight_rows(conn, profile_id)
    w = nearest_weight_kg(rows, d.get("date"))
    if w is None:
        return
    lean, fat = compute_lean_fat_kg(w, _floaty(bf))
    d["lean_mass"] = lean
    d["fat_mass"] = fat


def enrich_caliper_row_for_response(row: Dict[str, Any], weight_rows: List[Dict[str, Any]]) -> None:
    """Fill missing lean_mass / fat_mass for API responses (read path)."""
    bf = row.get("body_fat_pct")
    if bf is None:
        return
    if row.get("lean_mass") is not None and row.get("fat_mass") is not None:
        return
    w = nearest_weight_kg(weight_rows, row.get("date"))
    if w is None:
        return
    lean, fat = compute_lean_fat_kg(w, _floaty(bf))
    row["lean_mass"] = lean
    row["fat_mass"] = fat