from dataclasses import dataclass
from typing import Dict, Any, List
import re

REQ = [
    "loss_of_control_risk",
    "recovery_window_seconds",
    "recommended_adaptive_action",
    "intervention_priority",
    "recovery_confidence",
    "post_action_stability_expectation",
]

PRIORITY = ["low","medium","high","urgent"]

@dataclass
class ScoreResult:
    score: float
    details: Dict[str, Any]

def _f(p: str, key: str):
    m = re.search(rf"{key}\s*[:=]\s*(0\.\d+|1\.0)\b", p)
    return float(m.group(1)) if m else None

def _int(p: str, key: str):
    m = re.search(rf"{key}\s*[:=]\s*(\d+)", p)
    return int(m.group(1)) if m else None

def score(sample: Dict[str, Any], prediction: str) -> ScoreResult:
    p = (prediction or "").lower()
    words_ok = len(p.split()) <= 950

    hits = sum(1 for k in REQ if k in p)

    risk = _f(p,"loss_of_control_risk")
    win = _int(p,"recovery_window_seconds")
    conf = _f(p,"recovery_confidence")

    numeric_ok = int(
        risk is not None and 0<=risk<=1 and
        win is not None and 0<=win<=3600 and
        conf is not None and 0<=conf<=1
    )

    pr_ok = int("intervention_priority" in p and any(x in p for x in PRIORITY))
    action_ok = int("recommended_adaptive_action" in p)
    expect_ok = int("post_action_stability_expectation" in p)

    raw = (
        0.18 * int(words_ok) +
        0.40 * (hits/len(REQ)) +
        0.22 * numeric_ok +
        0.08 * pr_ok +
        0.06 * action_ok +
        0.06 * expect_ok
    )

    return ScoreResult(score=min(1.0,raw), details={"id": sample.get("id"),"hits":hits})

def aggregate(results: List[ScoreResult]) -> Dict[str, Any]:
    if not results:
        return {"mean":0.0,"n":0}
    return {"mean":sum(r.score for r in results)/len(results),"n":len(results)}