"""Predict a failure label with the published dependency-free model."""

from __future__ import annotations

import json
import math
import re
import sys
from collections import Counter
from pathlib import Path

TOKEN_RE = re.compile(r"[A-Za-zÀ-ÿ0-9_]+", re.UNICODE)


def tokens(text: str) -> list[str]:
    words = [word.casefold() for word in TOKEN_RE.findall(text)]
    return words + [f"{words[i]}__{words[i + 1]}" for i in range(len(words) - 1)]


def predict(model: dict, text: str) -> str:
    query = Counter(tokens(text))
    labels = model["labels"]
    vocab_size = len(model["vocabulary"])
    total = sum(model["class_counts"].values())
    best = (None, float("-inf"))
    for label in labels:
        score = math.log(model["class_counts"][label] / total)
        denominator = model["token_totals"][label] + vocab_size
        for token, count in query.items():
            score += count * math.log((model["token_counts"][label].get(token, 0) + 1) / denominator)
        if score > best[1]:
            best = (label, score)
    return best[0]


if __name__ == "__main__":
    model = json.loads((Path(__file__).parent / "model.json").read_text(encoding="utf-8"))
    text = " ".join(sys.argv[1:])
    if not text:
        raise SystemExit("usage: python predict.py 'texto da trajetória'")
    print(predict(model, text))