evaluation/weighted_score.py

"""
Weighted Coverage Score Calculator.

Computes weighted coverage scores from Likert (0-5) evaluations.

Formula:
    score = Sum(item_scores) / (5.0 * count_items)
    Result is 0.0 to 1.0  (multiply by 100 for percentage)

Adapted from calculate_weighted_coverage.py.
"""

from __future__ import annotations

from typing import Any, Dict, List, Optional


def calculate_score(coverage_data: Optional[Dict[str, Any]]) -> float:
    """
    Calculate the weighted coverage score from Likert evaluations.

    Args:
        coverage_data: Dict with "evaluations" key containing list of
                       evaluation dicts, each with a "score" field (0-5).

    Returns:
        Float between 0.0 and 1.0 representing the weighted coverage score.
        Returns 0.0 if input is empty or has no valid evaluations.
    """
    if not coverage_data:
        return 0.0

    evaluations = coverage_data.get("evaluations", [])
    if not evaluations:
        return 0.0

    total_score = 0
    count = 0

    for item in evaluations:
        s = item.get("score", 0)
        if isinstance(s, (int, float)):
            total_score += s
            count += 1

    if count == 0:
        return 0.0

    # Max score per item is 5
    max_possible = count * 5.0
    return total_score / max_possible


def calculate_score_percentage(coverage_data: Optional[Dict[str, Any]]) -> float:
    """
    Calculate the weighted coverage score as a percentage (0-100).

    Args:
        coverage_data: Dict with "evaluations" key.

    Returns:
        Float between 0.0 and 100.0.
    """
    return calculate_score(coverage_data) * 100.0


def aggregate_scores(scores: List[float]) -> float:
    """
    Compute the average of a list of scores.

    Args:
        scores: List of float scores.

    Returns:
        Average score, or 0.0 if empty.
    """
    if not scores:
        return 0.0
    return sum(scores) / len(scores)


def find_coverage_keys(data: List[Dict]) -> List[str]:
    """Find all keys ending with _coverage_scores in the data."""
    keys = set()
    for item in data:
        for key in item.keys():
            if key.endswith("_coverage_scores"):
                keys.add(key)
    return sorted(list(keys))