yt-shorts-downloader/src/yt_shorts_downloader/session.py

from __future__ import annotations

import time
from pathlib import Path

from .models import SessionCookie, SessionValidation

_SESSION_DOMAIN_SUFFIXES = ("youtube.com", "google.com")
_AUTH_COOKIE_NAMES = {
    "SID",
    "HSID",
    "SSID",
    "APISID",
    "SAPISID",
    "__Secure-1PSID",
    "__Secure-3PSID",
    "LOGIN_INFO",
}
_AUTH_COOKIE_ANCHORS = {
    "SID",
    "SAPISID",
    "__Secure-1PSID",
    "__Secure-3PSID",
    "LOGIN_INFO",
}


def validate_session_file(path: Path) -> SessionValidation:
    normalized_path = path.expanduser().resolve()
    if not normalized_path.exists():
        return SessionValidation(
            exists=False,
            structurally_valid=False,
            fresh=False,
            is_usable=False,
            message=f"Файл сессии не найден: {normalized_path}",
        )

    try:
        cookies = _parse_session_file(normalized_path)
    except ValueError as exc:
        return SessionValidation(
            exists=True,
            structurally_valid=False,
            fresh=False,
            is_usable=False,
            message=f"Файл сессии не прошёл проверку формата: {exc}",
        )

    if not cookies:
        return SessionValidation(
            exists=True,
            structurally_valid=False,
            fresh=False,
            is_usable=False,
            message="Файл сессии пустой",
        )

    relevant_cookies = [
        cookie
        for cookie in cookies
        if _matches_domain(cookie.domain, _SESSION_DOMAIN_SUFFIXES)
    ]
    if not relevant_cookies:
        return SessionValidation(
            exists=True,
            structurally_valid=False,
            fresh=False,
            is_usable=False,
            message="В файле сессии нет записей для YouTube или Google",
        )

    now = int(time.time())
    fresh_auth_names: set[str] = set()
    soonest_expiry: int | None = None

    for cookie in relevant_cookies:
        if cookie.name not in _AUTH_COOKIE_NAMES:
            continue
        if cookie.expires != 0 and cookie.expires <= now:
            continue

        fresh_auth_names.add(cookie.name)
        if cookie.expires > 0 and (
            soonest_expiry is None or cookie.expires < soonest_expiry
        ):
            soonest_expiry = cookie.expires

    anchor_count = len(fresh_auth_names & _AUTH_COOKIE_ANCHORS)
    fresh = len(fresh_auth_names) >= 3 and anchor_count >= 1
    if not fresh:
        return SessionValidation(
            exists=True,
            structurally_valid=True,
            fresh=False,
            is_usable=False,
            message=(
                "Файл сессии в корректном формате, но в нём нет достаточного набора "
                "актуальных YouTube auth-cookie"
            ),
        )

    message = (
        "Файл сессии валиден и выглядит актуальным для YouTube: "
        f"найдено {len(fresh_auth_names)} свежих auth-cookie"
    )
    if soonest_expiry is not None:
        hours_left = max((soonest_expiry - now) // 3600, 0)
        message += f", ближайшее истечение примерно через {hours_left} ч"

    return SessionValidation(
        exists=True,
        structurally_valid=True,
        fresh=True,
        is_usable=True,
        message=message,
    )


def _parse_session_file(path: Path) -> list[SessionCookie]:
    cookies: list[SessionCookie] = []

    with path.open("r", encoding="utf-8") as handle:
        for line_number, raw_line in enumerate(handle, start=1):
            stripped = raw_line.strip()
            if not stripped:
                continue

            is_httponly_cookie = stripped.startswith("#HttpOnly_")
            if stripped.startswith("#") and not is_httponly_cookie:
                continue

            if is_httponly_cookie:
                stripped = stripped.removeprefix("#HttpOnly_")

            parts = stripped.split("\t")
            if len(parts) != 7:
                raise ValueError(
                    "Строка "
                    f"{line_number}: ожидалось 7 колонок Netscape cookie file, "
                    f"получено {len(parts)}"
                )

            (
                domain,
                include_subdomains,
                cookie_path,
                secure,
                expires,
                name,
                value,
            ) = parts

            try:
                expires_at = int(expires)
            except ValueError as exc:
                raise ValueError(
                    f"Строка {line_number}: expires должен быть UNIX timestamp"
                ) from exc

            cookies.append(
                SessionCookie(
                    domain=domain,
                    include_subdomains=include_subdomains.upper() == "TRUE",
                    path=cookie_path,
                    secure=secure.upper() == "TRUE",
                    expires=expires_at,
                    name=name,
                    value=value,
                )
            )

    return cookies


def _matches_domain(domain: str, suffixes: tuple[str, ...]) -> bool:
    normalized_domain = _normalize_domain(domain)
    return any(
        normalized_domain == suffix or normalized_domain.endswith(f".{suffix}")
        for suffix in suffixes
    )


def _normalize_domain(domain: str) -> str:
    return domain.removeprefix("#HttpOnly_").lstrip(".").lower()