"""OSS/URL CSV source (v2).

- Validates incoming URL to reduce SSRF risk (allowlist + IP checks)
- Downloads CSV to a local temporary file for analysis

This module is intentionally small and dependency-light.
"""

from __future__ import annotations

import ipaddress
import logging
import os
import socket
import tempfile
from dataclasses import dataclass
from typing import Optional
from urllib.parse import urlsplit

import requests

from app.core.config import settings

logger = logging.getLogger(__name__)


@dataclass(frozen=True)
class DownloadedCsv:
    local_path: str
    source_host: str
    source_name: str
    etag: Optional[str] = None
    last_modified: Optional[str] = None


class UrlValidationError(ValueError):
    pass


def _is_ip_allowed(ip_str: str) -> bool:
    ip = ipaddress.ip_address(ip_str)

    if settings.V2_ALLOW_PRIVATE_NETWORKS:
        return True

    # Block loopback/link-local/private/multicast/unspecified/reserved
    if (
        ip.is_loopback
        or ip.is_private
        or ip.is_link_local
        or ip.is_multicast
        or ip.is_unspecified
        or ip.is_reserved
    ):
        return False

    return True


def validate_source_url(source_url: str) -> tuple[str, str]:
    """Validate URL and return (host, source_name)."""

    if not source_url or not isinstance(source_url, str):
        raise UrlValidationError("source_url 不能为空")

    parts = urlsplit(source_url)

    if parts.scheme not in {"https", "http"}:
        raise UrlValidationError("仅支持 http/https URL")

    if parts.scheme == "http" and not settings.V2_ALLOW_HTTP:
        raise UrlValidationError("不允许 http；请使用 https 或开启 V2_ALLOW_HTTP")

    if not parts.netloc:
        raise UrlValidationError("URL 缺少 host")

    # Disallow URLs with userinfo
    if "@" in parts.netloc:
        raise UrlValidationError("URL 不允许包含用户名/密码")

    host = parts.hostname
    if not host:
        raise UrlValidationError("无法解析 URL host")

    # Optional allowlist
    if settings.V2_ALLOWED_HOSTS:
        allowed = {h.lower() for h in settings.V2_ALLOWED_HOSTS}
        if host.lower() not in allowed:
            raise UrlValidationError(f"host 不在白名单: {host}")

    # Resolve host -> IP and block private/loopback, unless explicitly allowed.
    try:
        addr_info = socket.getaddrinfo(host, None)
    except socket.gaierror as e:
        raise UrlValidationError(f"DNS 解析失败: {host} ({e})") from e

    for family, _type, _proto, _canonname, sockaddr in addr_info:
        ip_str = None
        if family == socket.AF_INET:
            ip_str = str(sockaddr[0])
        elif family == socket.AF_INET6:
            ip_str = str(sockaddr[0])
        if ip_str and not _is_ip_allowed(ip_str):
            raise UrlValidationError(f"host 解析到不允许的 IP: {ip_str}")

    source_name = os.path.basename(parts.path) or "data.csv"
    return host, source_name


def download_csv_to_tempfile(source_url: str, *, suffix: str = ".csv") -> DownloadedCsv:
    """Download URL content to a temp file and return local path + meta."""

    host, source_name = validate_source_url(source_url)

    # Create temp file inside configured TEMP_DIR for easier ops/observability
    settings.TEMP_DIR.mkdir(exist_ok=True)
    tmp = tempfile.NamedTemporaryFile(
        mode="wb",
        suffix=suffix,
        dir=str(settings.TEMP_DIR),
        delete=False,
    )

    try:
        timeout = (settings.V2_CONNECT_TIMEOUT_SECONDS, settings.V2_DOWNLOAD_TIMEOUT_SECONDS)
        with requests.get(source_url, stream=True, timeout=timeout) as resp:
            resp.raise_for_status()
            etag = resp.headers.get("ETag")
            last_modified = resp.headers.get("Last-Modified")

            for chunk in resp.iter_content(chunk_size=1024 * 1024):
                if not chunk:
                    continue
                tmp.write(chunk)

        tmp.flush()
        tmp.close()

        if os.path.getsize(tmp.name) <= 0:
            raise UrlValidationError("下载内容为空")

        return DownloadedCsv(
            local_path=tmp.name,
            source_host=host,
            source_name=source_name,
            etag=etag,
            last_modified=last_modified,
        )

    except Exception:
        try:
            tmp.close()
        except Exception:
            pass
        try:
            os.unlink(tmp.name)
        except Exception:
            pass
        raise