mitai-jinkendo/backend/calculations/body_metrics.py

"""
Body Metrics Calculation Engine

Implements K1-K5 from visualization concept:
- K1: Weight trend + goal projection
- K2: Weight/FM/LBM multi-line chart
- K3: Circumference panel
- K4: Recomposition detector
- K5: Body progress score (goal-mode dependent)

All calculations include data quality/confidence assessment.
"""
from datetime import datetime, timedelta
from typing import Optional, Dict, Tuple
import statistics

from db import get_db, get_cursor


# ============================================================================
# K1: Weight Trend Calculations
# ============================================================================

def calculate_weight_7d_median(profile_id: str) -> Optional[float]:
    """Calculate 7-day median weight (reduces daily noise)"""
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute("""
            SELECT weight
            FROM weight_log
            WHERE profile_id = %s
              AND date >= CURRENT_DATE - INTERVAL '7 days'
            ORDER BY date DESC
        """, (profile_id,))

        weights = [row['weight'] for row in cur.fetchall()]

        if len(weights) < 4:  # Need at least 4 measurements
            return None

        return round(statistics.median(weights), 1)


def calculate_weight_28d_slope(profile_id: str) -> Optional[float]:
    """Calculate 28-day weight slope (kg/day)"""
    return _calculate_weight_slope(profile_id, days=28)


def calculate_weight_90d_slope(profile_id: str) -> Optional[float]:
    """Calculate 90-day weight slope (kg/day)"""
    return _calculate_weight_slope(profile_id, days=90)


def _calculate_weight_slope(profile_id: str, days: int) -> Optional[float]:
    """
    Calculate weight slope using linear regression
    Returns kg/day (negative = weight loss)
    """
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute("""
            SELECT date, weight
            FROM weight_log
            WHERE profile_id = %s
              AND date >= CURRENT_DATE - INTERVAL '%s days'
            ORDER BY date
        """, (profile_id, days))

        data = [(row['date'], row['weight']) for row in cur.fetchall()]

        # Need minimum data points based on period
        min_points = max(18, int(days * 0.6))  # 60% coverage
        if len(data) < min_points:
            return None

        # Convert dates to days since start
        start_date = data[0][0]
        x_values = [(date - start_date).days for date, _ in data]
        y_values = [weight for _, weight in data]

        # Linear regression
        n = len(data)
        x_mean = sum(x_values) / n
        y_mean = sum(y_values) / n

        numerator = sum((x - x_mean) * (y - y_mean) for x, y in zip(x_values, y_values))
        denominator = sum((x - x_mean) ** 2 for x in x_values)

        if denominator == 0:
            return None

        slope = numerator / denominator
        return round(slope, 4)  # kg/day


def calculate_goal_projection_date(profile_id: str, goal_id: str) -> Optional[str]:
    """
    Calculate projected date to reach goal based on 28d trend
    Returns ISO date string or None if unrealistic
    """
    from goal_utils import get_goal_by_id

    goal = get_goal_by_id(goal_id)
    if not goal or goal['goal_type'] != 'weight':
        return None

    slope = calculate_weight_28d_slope(profile_id)
    if not slope or slope == 0:
        return None

    current = goal['current_value']
    target = goal['target_value']
    remaining = target - current

    days_needed = remaining / slope

    # Unrealistic if >2 years or negative
    if days_needed < 0 or days_needed > 730:
        return None

    projection_date = datetime.now().date() + timedelta(days=int(days_needed))
    return projection_date.isoformat()


def calculate_goal_progress_pct(current: float, target: float, start: float) -> int:
    """
    Calculate goal progress percentage
    Returns 0-100 (can exceed 100 if target surpassed)
    """
    if start == target:
        return 100 if current == target else 0

    progress = ((current - start) / (target - start)) * 100
    return max(0, min(100, int(progress)))


# ============================================================================
# K2: Fat Mass / Lean Mass Calculations
# ============================================================================

def calculate_fm_28d_change(profile_id: str) -> Optional[float]:
    """Calculate 28-day fat mass change (kg)"""
    return _calculate_body_composition_change(profile_id, 'fm', 28)


def calculate_lbm_28d_change(profile_id: str) -> Optional[float]:
    """Calculate 28-day lean body mass change (kg)"""
    return _calculate_body_composition_change(profile_id, 'lbm', 28)


def _calculate_body_composition_change(profile_id: str, metric: str, days: int) -> Optional[float]:
    """
    Calculate change in body composition over period
    metric: 'fm' (fat mass) or 'lbm' (lean mass)
    """
    with get_db() as conn:
        cur = get_cursor(conn)

        # Get weight and caliper measurements
        cur.execute("""
            SELECT w.date, w.weight, c.body_fat_pct
            FROM weight_log w
            LEFT JOIN caliper_log c ON w.profile_id = c.profile_id
                AND w.date = c.date
            WHERE w.profile_id = %s
              AND w.date >= CURRENT_DATE - INTERVAL '%s days'
            ORDER BY w.date DESC
        """, (profile_id, days))

        data = [
            {
                'date': row['date'],
                'weight': row['weight'],
                'bf_pct': row['body_fat_pct']
            }
            for row in cur.fetchall()
            if row['body_fat_pct'] is not None  # Need BF% for composition
        ]

        if len(data) < 2:
            return None

        # Most recent and oldest measurement
        recent = data[0]
        oldest = data[-1]

        # Calculate FM and LBM
        recent_fm = recent['weight'] * (recent['bf_pct'] / 100)
        recent_lbm = recent['weight'] - recent_fm

        oldest_fm = oldest['weight'] * (oldest['bf_pct'] / 100)
        oldest_lbm = oldest['weight'] - oldest_fm

        if metric == 'fm':
            change = recent_fm - oldest_fm
        else:  # lbm
            change = recent_lbm - oldest_lbm

        return round(change, 2)


# ============================================================================
# K3: Circumference Calculations
# ============================================================================

def calculate_waist_28d_delta(profile_id: str) -> Optional[float]:
    """Calculate 28-day waist circumference change (cm)"""
    return _calculate_circumference_delta(profile_id, 'c_waist', 28)


def calculate_hip_28d_delta(profile_id: str) -> Optional[float]:
    """Calculate 28-day hip circumference change (cm)"""
    return _calculate_circumference_delta(profile_id, 'c_hip', 28)


def calculate_chest_28d_delta(profile_id: str) -> Optional[float]:
    """Calculate 28-day chest circumference change (cm)"""
    return _calculate_circumference_delta(profile_id, 'c_chest', 28)


def calculate_arm_28d_delta(profile_id: str) -> Optional[float]:
    """Calculate 28-day arm circumference change (cm)"""
    return _calculate_circumference_delta(profile_id, 'c_arm', 28)


def calculate_thigh_28d_delta(profile_id: str) -> Optional[float]:
    """Calculate 28-day thigh circumference change (cm, average of L/R)"""
    left = _calculate_circumference_delta(profile_id, 'c_thigh', 28)
    right = _calculate_circumference_delta(profile_id, 'c_thigh_r', 28)

    if left is None or right is None:
        return None

    return round((left + right) / 2, 1)


def _calculate_circumference_delta(profile_id: str, column: str, days: int) -> Optional[float]:
    """Calculate change in circumference measurement"""
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute(f"""
            SELECT {column}
            FROM circumference_log
            WHERE profile_id = %s
              AND date >= CURRENT_DATE - INTERVAL '%s days'
              AND {column} IS NOT NULL
            ORDER BY date DESC
            LIMIT 1
        """, (profile_id, days))

        recent = cur.fetchone()
        if not recent:
            return None

        cur.execute(f"""
            SELECT {column}
            FROM circumference_log
            WHERE profile_id = %s
              AND date < CURRENT_DATE - INTERVAL '%s days'
              AND {column} IS NOT NULL
            ORDER BY date DESC
            LIMIT 1
        """, (profile_id, days))

        oldest = cur.fetchone()
        if not oldest:
            return None

        change = recent[column] - oldest[column]
        return round(change, 1)


def calculate_waist_hip_ratio(profile_id: str) -> Optional[float]:
    """Calculate current waist-to-hip ratio"""
    with get_db() as conn:
        cur = get_cursor(conn)
        cur.execute("""
            SELECT c_waist, c_hip
            FROM circumference_log
            WHERE profile_id = %s
              AND c_waist IS NOT NULL
              AND c_hip IS NOT NULL
            ORDER BY date DESC
            LIMIT 1
        """, (profile_id,))

        row = cur.fetchone()
        if not row:
            return None

        ratio = row['c_waist'] / row['c_hip']
        return round(ratio, 3)


# ============================================================================
# K4: Recomposition Detector
# ============================================================================

def calculate_recomposition_quadrant(profile_id: str) -> Optional[str]:
    """
    Determine recomposition quadrant based on 28d changes:
    - optimal: FM down, LBM up
    - cut_with_risk: FM down, LBM down
    - bulk: FM up, LBM up
    - unfavorable: FM up, LBM down
    """
    fm_change = calculate_fm_28d_change(profile_id)
    lbm_change = calculate_lbm_28d_change(profile_id)

    if fm_change is None or lbm_change is None:
        return None

    if fm_change < 0 and lbm_change > 0:
        return "optimal"
    elif fm_change < 0 and lbm_change < 0:
        return "cut_with_risk"
    elif fm_change > 0 and lbm_change > 0:
        return "bulk"
    else:  # fm_change > 0 and lbm_change < 0
        return "unfavorable"


# ============================================================================
# K5: Body Progress Score (Dynamic Focus Areas)
# ============================================================================

def calculate_body_progress_score(profile_id: str, focus_weights: Optional[Dict] = None) -> Optional[int]:
    """
    Calculate body progress score (0-100) weighted by user's focus areas

    Components:
    - Weight trend alignment with goals
    - FM/LBM changes (recomposition quality)
    - Circumference changes (especially waist)
    - Goal progress percentage

    Weighted dynamically based on user's focus area priorities
    """
    if focus_weights is None:
        from calculations.scores import get_user_focus_weights
        focus_weights = get_user_focus_weights(profile_id)

    # Get all body-related focus area weights
    body_weight = focus_weights.get('körpergewicht', 0)
    body_fat_weight = focus_weights.get('körperfett', 0)
    muscle_weight = focus_weights.get('muskelmasse', 0)

    total_body_weight = body_weight + body_fat_weight + muscle_weight

    if total_body_weight == 0:
        return None  # No body-related goals

    # Calculate component scores (0-100)
    components = []

    # Weight trend component (if weight goal active)
    if body_weight > 0:
        weight_score = _score_weight_trend(profile_id)
        if weight_score is not None:
            components.append(('weight', weight_score, body_weight))

    # Body composition component (if BF% or LBM goal active)
    if body_fat_weight > 0 or muscle_weight > 0:
        comp_score = _score_body_composition(profile_id)
        if comp_score is not None:
            components.append(('composition', comp_score, body_fat_weight + muscle_weight))

    # Waist circumference component (proxy for health)
    waist_score = _score_waist_trend(profile_id)
    if waist_score is not None:
        # Waist gets 20% base weight + bonus from BF% goals
        waist_weight = 20 + (body_fat_weight * 0.3)
        components.append(('waist', waist_score, waist_weight))

    if not components:
        return None

    # Weighted average
    total_score = sum(score * weight for _, score, weight in components)
    total_weight = sum(weight for _, _, weight in components)

    return int(total_score / total_weight)


def _score_weight_trend(profile_id: str) -> Optional[int]:
    """Score weight trend alignment with goals (0-100)"""
    from goal_utils import get_goals_by_type

    goals = get_goals_by_type(profile_id, 'weight')
    if not goals:
        return None

    # Use primary or first active goal
    goal = next((g for g in goals if g.get('is_primary')), goals[0])

    current = goal.get('current_value')
    target = goal.get('target_value')
    start = goal.get('start_value', current)

    if None in [current, target, start]:
        return None

    # Progress percentage
    progress_pct = calculate_goal_progress_pct(current, target, start)

    # Bonus/penalty based on trend
    slope = calculate_weight_28d_slope(profile_id)
    if slope is not None:
        desired_direction = -1 if target < start else 1
        actual_direction = -1 if slope < 0 else 1

        if desired_direction == actual_direction:
            # Moving in right direction
            score = min(100, progress_pct + 10)
        else:
            # Moving in wrong direction
            score = max(0, progress_pct - 20)
    else:
        score = progress_pct

    return int(score)


def _score_body_composition(profile_id: str) -> Optional[int]:
    """Score body composition changes (0-100)"""
    fm_change = calculate_fm_28d_change(profile_id)
    lbm_change = calculate_lbm_28d_change(profile_id)

    if fm_change is None or lbm_change is None:
        return None

    quadrant = calculate_recomposition_quadrant(profile_id)

    # Scoring by quadrant
    if quadrant == "optimal":
        return 100
    elif quadrant == "cut_with_risk":
        # Penalty proportional to LBM loss
        penalty = min(30, abs(lbm_change) * 15)
        return max(50, 80 - int(penalty))
    elif quadrant == "bulk":
        # Score based on FM/LBM ratio
        if lbm_change > 0 and fm_change > 0:
            ratio = lbm_change / fm_change
            if ratio >= 3:  # 3:1 LBM:FM = excellent bulk
                return 90
            elif ratio >= 2:
                return 75
            elif ratio >= 1:
                return 60
            else:
                return 45
        return 60
    else:  # unfavorable
        return 20


def _score_waist_trend(profile_id: str) -> Optional[int]:
    """Score waist circumference trend (0-100)"""
    delta = calculate_waist_28d_delta(profile_id)

    if delta is None:
        return None

    # Waist reduction is almost always positive
    if delta <= -3:  # >3cm reduction
        return 100
    elif delta <= -2:
        return 90
    elif delta <= -1:
        return 80
    elif delta <= 0:
        return 70
    elif delta <= 1:
        return 55
    elif delta <= 2:
        return 40
    else:  # >2cm increase
        return 20


# ============================================================================
# Data Quality Assessment
# ============================================================================

def calculate_body_data_quality(profile_id: str) -> Dict[str, any]:
    """
    Assess data quality for body metrics
    Returns dict with quality score and details
    """
    with get_db() as conn:
        cur = get_cursor(conn)

        # Weight measurement frequency (last 28 days)
        cur.execute("""
            SELECT COUNT(*) as count
            FROM weight_log
            WHERE profile_id = %s
              AND date >= CURRENT_DATE - INTERVAL '28 days'
        """, (profile_id,))
        weight_count = cur.fetchone()['count']

        # Caliper measurement frequency (last 28 days)
        cur.execute("""
            SELECT COUNT(*) as count
            FROM caliper_log
            WHERE profile_id = %s
              AND date >= CURRENT_DATE - INTERVAL '28 days'
        """, (profile_id,))
        caliper_count = cur.fetchone()['count']

        # Circumference measurement frequency (last 28 days)
        cur.execute("""
            SELECT COUNT(*) as count
            FROM circumference_log
            WHERE profile_id = %s
              AND date >= CURRENT_DATE - INTERVAL '28 days'
        """, (profile_id,))
        circ_count = cur.fetchone()['count']

    # Score components
    weight_score = min(100, (weight_count / 18) * 100)  # 18 = ~65% of 28 days
    caliper_score = min(100, (caliper_count / 4) * 100)  # 4 = weekly
    circ_score = min(100, (circ_count / 4) * 100)

    # Overall score (weight 50%, caliper 30%, circ 20%)
    overall_score = int(
        weight_score * 0.5 +
        caliper_score * 0.3 +
        circ_score * 0.2
    )

    # Confidence level
    if overall_score >= 80:
        confidence = "high"
    elif overall_score >= 60:
        confidence = "medium"
    else:
        confidence = "low"

    return {
        "overall_score": overall_score,
        "confidence": confidence,
        "measurements": {
            "weight_28d": weight_count,
            "caliper_28d": caliper_count,
            "circumference_28d": circ_count
        },
        "component_scores": {
            "weight": int(weight_score),
            "caliper": int(caliper_score),
            "circumference": int(circ_score)
        }
    }