新增功能: - execution_log 模块:自动记录引擎调用耗时、成功/失败、错误详情 - 所有搜索引擎自动跟踪每次调用的性能指标 - CLI 搜索后自动显示执行摘要 - health 命令使用 ParallelSearch 检测所有引擎连通性 - 新增 log 命令查看历史执行日志
67 lines
2.2 KiB
Python
67 lines
2.2 KiB
Python
"""Sogou HTML scraping engine — free, good Chinese results."""
|
|
|
|
from __future__ import annotations
|
|
|
|
import time
|
|
import logging
|
|
from urllib.parse import quote_plus
|
|
|
|
import httpx
|
|
from bs4 import BeautifulSoup
|
|
|
|
from scripts.engines.base import BaseEngine, EngineResult
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
HEADERS = {
|
|
"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
|
|
"(KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36",
|
|
"Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
|
|
"Accept-Language": "zh-CN,zh;q=0.9",
|
|
}
|
|
|
|
|
|
class SogouScraper(BaseEngine):
|
|
"""Sogou web scraping — best for Chinese content."""
|
|
|
|
name = "sogou"
|
|
|
|
async def search(self, query: str, max_results: int = 10,
|
|
**kwargs) -> list[EngineResult]:
|
|
t0 = time.time()
|
|
url = f"https://www.sogou.com/web?query={quote_plus(query)}"
|
|
|
|
try:
|
|
async with httpx.AsyncClient(timeout=self.timeout, follow_redirects=True) as client:
|
|
resp = await client.get(url, headers=HEADERS)
|
|
resp.raise_for_status()
|
|
elapsed = time.time() - t0
|
|
results = self._parse(resp.text, max_results)
|
|
logger.info(f"[sogou] {len(results)} results in {elapsed:.2f}s")
|
|
return results
|
|
except Exception as e:
|
|
logger.warning(f"[sogou] failed: {e}")
|
|
return []
|
|
|
|
def _parse(self, html: str, max_results: int) -> list[EngineResult]:
|
|
soup = BeautifulSoup(html, "html.parser")
|
|
results = []
|
|
|
|
for item in soup.select("div.results > div"):
|
|
title_el = item.select_one("h3 a, a.account")
|
|
if not title_el:
|
|
continue
|
|
title = title_el.get_text(strip=True)
|
|
href = title_el.get("href", "")
|
|
snippet_el = item.select_one("p.str_info, div.str-text, .star-wiki")
|
|
snippet = snippet_el.get_text(strip=True) if snippet_el else ""
|
|
|
|
if title and href and not href.startswith("/"):
|
|
results.append(EngineResult(
|
|
title=title, url=href, snippet=snippet, source="sogou",
|
|
))
|
|
if len(results) >= max_results:
|
|
break
|
|
|
|
return results
|