# -*- coding: UTF-8 -*-
'''
@Project :domainScanDemo
@File    :c360.py
@IDE     :PyCharm
@Author  :梦伴
@Date    :2026/4/2 15:57
@explain : 360搜索敏感词检测工具 - 检测域名是否包含敏感词
'''

import re  # 正则表达式模块
import os
import threading
import time
import requests  # HTTP请求库
from loguru import logger  # 日志记录
from requests.adapters import HTTPAdapter
from typing import List, Tuple, Optional, Dict  # 类型提示


# 常量定义
SO_SEARCH_URL = 'https://www.so.com/s'  # 360搜索URL
SO_REFERER_TEMPLATE = 'https://www.so.com/s?ie=utf-8&q=site%3A{domain}'  # Referer模板
SEARCH_PATTERN = r'target="_blank">([^<]+)</a></h3>'  # 搜索结果匹配模式
BLOCKED_CODE = 3  # 拦截状态码
_SESSION_LOCAL = threading.local()
SO_TIMEOUT_PROXY = max(
    1.0,
    float(os.getenv("DOMAINCHECK_360_TIMEOUT_PROXY", "1.4") or 1.4),
)
SO_TIMEOUT_DIRECT = max(
    1.2,
    float(os.getenv("DOMAINCHECK_360_TIMEOUT_DIRECT", "1.8") or 1.8),
)


def _resolve_360_timeout(proxies: Optional[Dict] = None, budget_seconds: Optional[float] = None) -> float:
    timeout = float(SO_TIMEOUT_PROXY if proxies else SO_TIMEOUT_DIRECT)
    if budget_seconds not in (None, "", 0, "0"):
        timeout = min(timeout, max(0.6, float(budget_seconds or 0.0)))
    return max(0.6, timeout)


def _get_session():
    session = getattr(_SESSION_LOCAL, "session", None)
    if session is not None:
        return session
    session = requests.Session()
    session.mount("http://", HTTPAdapter(pool_connections=64, pool_maxsize=128, max_retries=0))
    session.mount("https://", HTTPAdapter(pool_connections=64, pool_maxsize=128, max_retries=0))
    _SESSION_LOCAL.session = session
    return session


def check_domain(
    domain: str,
    sensitive_words: Optional[List[str]] = None,
    proxies: Optional[Dict] = None,
    budget_seconds: Optional[float] = None,
) -> Tuple[bool, str]:
    """
    检查域名是否包含敏感词

    通过360搜索获取域名的搜索结果，然后检查搜索结果中是否包含敏感词

    Args:
        domain: 要检测的域名
        sensitive_words: 敏感词列表，如果为None则初始化为空列表
        proxies: 代理配置

    Returns:
        Tuple[bool, str]: (是否通过, 错误信息)
            - True, '': 通过检测，不包含敏感词
            - False, '敏感词：xxx': 包含敏感词，返回具体的敏感词
    """
    # 初始化敏感词列表
    if sensitive_words is None:
        sensitive_words = []

    # 构建请求头
    headers = {
        'Accept': '*/*',  # 接受所有类型
        'Accept-Language': 'zh-cn',  # 中文语言
        'Referer': SO_REFERER_TEMPLATE.format(domain=domain),  # 来源页面
        'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.87 Safari/537.36',  # 用户代理
        'Host': 'www.so.com',  # 主机名
    }

    # 构建查询参数
    params = {
        'ie': 'utf-8',  # 编码格式
        'q': f'site:{domain}',  # 搜索查询
    }

    try:
        session_started_at = time.perf_counter()
        logger.info(f'360阶段: domain={domain} | stage=session_init_start')
        session = _get_session()
        logger.info(
            f'360阶段: domain={domain} | stage=session_init_done '
            f'| elapsed_ms={int((time.perf_counter() - session_started_at) * 1000)}'
        )
        # 发送GET请求获取搜索结果
        request_started_at = time.perf_counter()
        logger.info(f'360阶段: domain={domain} | stage=request_start | proxy={"yes" if proxies else "no"}')
        response = session.get(
            SO_SEARCH_URL,
            params=params,
            headers=headers,
            proxies=proxies,
            timeout=_resolve_360_timeout(proxies, budget_seconds=budget_seconds),
        )
        logger.info(
            f'360阶段: domain={domain} | stage=request_done | status={response.status_code} '
            f'| elapsed_ms={int((time.perf_counter() - request_started_at) * 1000)}'
        )

        # 解析响应内容
        response_html = response.content.decode('utf-8', errors='ignore')

        # 使用正则表达式提取搜索结果
        search_results = re.findall(SEARCH_PATTERN, response_html)
        logger.info(f'360阶段: domain={domain} | stage=parsed | result_count={len(search_results)}')

        # 记录搜索结果数量
        logger.info(f'360搜索结果数量: {len(search_results)}')

        # 如果没有搜索结果，直接返回通过
        if not search_results:
            return True, ''

        # 检查每个搜索结果是否包含敏感词
        for result in search_results:
            for sensitive_word in sensitive_words:
                # 使用正则表达式检查是否包含敏感词
                if re.search(sensitive_word, result, re.IGNORECASE):
                    logger.warning(f'检测到敏感词: {sensitive_word} 在结果: {result}')
                    return False, f'检测到敏感词: {sensitive_word} 在结果: {result}'

        # 所有搜索结果都不包含敏感词，返回通过
        return True, ''

    except requests.exceptions.Timeout:
        return False, '360搜索请求超时'
    except requests.exceptions.RequestException as e:
        error_str = str(e)
        logger.debug(f'360搜索请求失败: {error_str}')
        # 检查是否是"Too many open files"错误
        if 'Too many open files' in error_str:
            return True, 'failure'
        return False, f'360搜索请求失败: {error_str}'
    except Exception as e:
        return False, f'未知错误: {str(e)}'


# if __name__ == '__main__':
#     # 测试用例
#     test_domain = 'niuniushouka.com'
#     test_sensitive_words = ['赌博', '色情', '暴力','收卡']
#
#     logger.info(f'开始检测域名: {test_domain}')
#     result = check_domain(test_domain, test_sensitive_words)
#     logger.info(f'检测结果: {result}')