grader.py

from runner import run_case


def compare_output(output, expected):
    return output == expected


def evaluate_task(task, solve_fn):
    cases = task["public_tests"] + task["hidden_tests"]
    score_per_case = 10
    max_score = len(cases) * score_per_case

    passed = 0
    failures = []
    total_runtime = 0.0

    for idx, case in enumerate(cases, start=1):
        result = run_case(solve_fn, task["name"], case["input"])
        total_runtime += result["runtime_ms"]

        if result["ok"] and compare_output(result["output"], case["expected"]):
            passed += 1
        else:
            failures.append(
                {
                    "case_id": idx,
                    "input": case["input"],
                    "expected": case["expected"],
                    "output": result["output"],
                    "error": result["error"] if not result["ok"] else "WrongAnswer",
                }
            )

    score = passed * score_per_case

    return {
        "task_name": task["name"],
        "score": score,
        "max_score": max_score,
        "passed": passed,
        "total": len(cases),
        "failures": failures,
        "avg_runtime_ms": total_runtime / len(cases) if cases else 0.0,
    }