sup_funcs.py

from __future__ import annotations

import re
from pathlib import Path

import pandas as pd


SEVERITY_TRANSLATIONS = {
    "Critical": "Критический",
    "High": "Высокий",
    "Medium": "Средний",
    "Low": "Низкий",
}


def upd_regs(frame: pd.DataFrame) -> pd.DataFrame:
    """Normalize RedCheck remediation text and extract KB identifiers."""

    result = frame.copy()

    def normalize(value: object) -> str:
        if not isinstance(value, str):
            return ""

        kb_matches = re.findall(r">([A-Z]{2}[0-9]{3,9})", value)
        if kb_matches:
            return "Необходимо установить обновления " + " ".join(kb_matches)

        # Legacy reports sometimes stored a URL inside an HTML anchor.
        href_match = re.search(r'href=["\']([^"\']+)["\']', value, flags=re.IGNORECASE)
        if href_match:
            return href_match.group(1)

        # Preserve readable text when it is not one of the known legacy patterns.
        return re.sub(r"<[^>]+>", "", value).strip()

    result["def_remediation"] = result["def_remediation"].map(normalize)
    return result


def ip2names(frame: pd.DataFrame, file_soot: str | Path | None) -> pd.DataFrame:
    """Replace address with a mapped network name when an optional CSV is provided."""

    result = frame.copy()
    mapping_path = Path(file_soot).expanduser() if file_soot else None

    if mapping_path:
        names = pd.read_csv(mapping_path, sep=";", encoding="utf-8")
        required = {"address", "name"}
        missing = required.difference(names.columns)
        if missing:
            raise ValueError(
                "В CSV соответствия отсутствуют колонки: " + ", ".join(sorted(missing))
            )

        names = names[["address", "name"]].drop_duplicates(subset=["address"], keep="last")
        result = result.merge(names, on="address", how="left")
        result["name"] = result["name"].fillna(result["address"])
    else:
        result["name"] = result["address"]

    keep_columns = [
        "name",
        "product",
        "title",
        "def_description",
        "def_severity",
        "def_remediation",
        "reference_source",
        "ref_id",
        "ref_url",
        "def_altx_id",
    ]
    return result[keep_columns]


def severity_rus(frame: pd.DataFrame) -> pd.DataFrame:
    result = frame.copy()
    result["def_severity"] = result["def_severity"].replace(SEVERITY_TRANSLATIONS)
    return result


def col_fstec(frame: pd.DataFrame) -> pd.DataFrame:
    result = frame.copy()
    result["BDU FSTEC"] = "-"
    result.loc[result["reference_source"] == "FSTEC", "BDU FSTEC"] = "+"
    return result


def group_name(frame: pd.DataFrame) -> pd.DataFrame:
    """Group one vulnerability across hosts while preserving report fields."""

    result = frame.copy()

    def host_list(values: pd.Series) -> str:
        unique = sorted({str(value) for value in values.dropna()})
        return str(unique)

    host_groups = (
        result.groupby("def_altx_id", dropna=False)["name"]
        .agg(host_list)
        .rename("_grouped_names")
    )

    result = result.drop_duplicates(subset=["def_altx_id"], keep="first")
    result = result.join(host_groups, on="def_altx_id")
    result["name"] = result.pop("_grouped_names")
    return result