实现浏览器方式互联网搜索功能
- 使用 agent-browser 浏览器自动化进行 Bing 搜索 - 解析 accessibility tree 提取搜索结果 - 自动获取每个结果的 URL - 支持设置最大结果数量
This commit is contained in:
+133
-5
@@ -4,6 +4,10 @@
|
|||||||
import requests
|
import requests
|
||||||
from bs4 import BeautifulSoup
|
from bs4 import BeautifulSoup
|
||||||
import json
|
import json
|
||||||
|
import subprocess
|
||||||
|
import os
|
||||||
|
import re
|
||||||
|
import urllib.parse
|
||||||
from datetime import datetime
|
from datetime import datetime
|
||||||
from config import Config
|
from config import Config
|
||||||
from models.database import db
|
from models.database import db
|
||||||
@@ -13,19 +17,143 @@ class SearchService:
|
|||||||
self.timeout = Config.SEARCH_TIMEOUT
|
self.timeout = Config.SEARCH_TIMEOUT
|
||||||
self.max_results = Config.SEARCH_MAX_RESULTS
|
self.max_results = Config.SEARCH_MAX_RESULTS
|
||||||
|
|
||||||
|
def _run_browser(self, *args, timeout=30000):
|
||||||
|
"""运行 agent-browser 命令"""
|
||||||
|
env = os.environ.copy()
|
||||||
|
env['XDG_RUNTIME_DIR'] = '/tmp/agent-browser-runtime'
|
||||||
|
os.makedirs(env['XDG_RUNTIME_DIR'], exist_ok=True)
|
||||||
|
|
||||||
|
cmd = ['agent-browser'] + list(args)
|
||||||
|
result = subprocess.run(
|
||||||
|
cmd,
|
||||||
|
capture_output=True,
|
||||||
|
text=True,
|
||||||
|
env=env,
|
||||||
|
timeout=timeout // 1000 + 5
|
||||||
|
)
|
||||||
|
return result.stdout, result.stderr, result.returncode
|
||||||
|
|
||||||
def search_internet(self, keyword, max_results=None):
|
def search_internet(self, keyword, max_results=None):
|
||||||
"""
|
"""
|
||||||
从互联网搜索(使用搜索API或爬虫)
|
从互联网搜索(使用 agent-browser 浏览器自动化)
|
||||||
这里暂时使用简单的搜索模拟
|
|
||||||
"""
|
"""
|
||||||
max_results = max_results or self.max_results
|
max_results = max_results or self.max_results
|
||||||
|
|
||||||
# TODO: 接入真实的搜索API(如Google Custom Search、Bing等)
|
|
||||||
# 这里先返回空列表,等待后续接入真实API
|
|
||||||
results = []
|
results = []
|
||||||
|
|
||||||
|
try:
|
||||||
|
# 1. 打开 Bing 搜索
|
||||||
|
encoded_keyword = urllib.parse.quote(keyword)
|
||||||
|
search_url = f"https://www.bing.com/search?q={encoded_keyword}"
|
||||||
|
|
||||||
|
stdout, stderr, code = self._run_browser('open', search_url, '--timeout', '20000')
|
||||||
|
if code != 0:
|
||||||
|
print(f"打开搜索页面失败: {stderr}")
|
||||||
|
return results
|
||||||
|
|
||||||
|
# 等待页面加载
|
||||||
|
stdout, stderr, code = self._run_browser('wait', '5000')
|
||||||
|
|
||||||
|
# 2. 获取搜索结果页面结构 (JSON 格式)
|
||||||
|
stdout, stderr, code = self._run_browser('snapshot', '--json', '--timeout', '30000')
|
||||||
|
if code != 0:
|
||||||
|
print(f"获取页面结构失败: {stderr}")
|
||||||
|
return results
|
||||||
|
|
||||||
|
# 3. 解析 JSON 提取搜索结果
|
||||||
|
try:
|
||||||
|
data = json.loads(stdout)
|
||||||
|
except json.JSONDecodeError:
|
||||||
|
print(f"解析 JSON 失败: {stdout[:500]}")
|
||||||
|
return results
|
||||||
|
|
||||||
|
# 4. 从 accessibility tree 中提取搜索结果
|
||||||
|
# Bing 搜索结果在 main[aria-label="搜索结果"] 区域内
|
||||||
|
results = self._parse_bing_results(data, max_results)
|
||||||
|
|
||||||
|
# 5. 关闭浏览器
|
||||||
|
self._run_browser('close')
|
||||||
|
|
||||||
|
except subprocess.TimeoutExpired:
|
||||||
|
print(f"搜索超时: {keyword}")
|
||||||
|
except Exception as e:
|
||||||
|
print(f"搜索出错: {str(e)}")
|
||||||
|
# 尝试关闭浏览器
|
||||||
|
try:
|
||||||
|
self._run_browser('close')
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
|
||||||
return results
|
return results
|
||||||
|
|
||||||
|
def _parse_bing_results(self, snapshot_data, max_results=10):
|
||||||
|
"""
|
||||||
|
从 Bing 搜索结果的 snapshot 中解析出标题和链接
|
||||||
|
|
||||||
|
snapshot_data 是 agent-browser snapshot --json 的输出
|
||||||
|
结构: {success, data: {snapshot: "文本格式的 accessibility tree"}, error}
|
||||||
|
"""
|
||||||
|
results = []
|
||||||
|
|
||||||
|
# 获取 snapshot 文本
|
||||||
|
snapshot = snapshot_data.get('data', {}).get('snapshot', '')
|
||||||
|
if not snapshot:
|
||||||
|
return results
|
||||||
|
|
||||||
|
# 解析 accessibility tree 文本
|
||||||
|
in_results = False
|
||||||
|
refs = [] # 存储 (title, ref) 元组
|
||||||
|
lines = snapshot.split('\n')
|
||||||
|
|
||||||
|
for i, line in enumerate(lines):
|
||||||
|
line = line.strip()
|
||||||
|
|
||||||
|
# 进入搜索结果区域
|
||||||
|
if 'main "搜索结果"' in line:
|
||||||
|
in_results = True
|
||||||
|
continue
|
||||||
|
|
||||||
|
# 离开搜索结果区域
|
||||||
|
if in_results and line.startswith('- ') and 'main' in line and '搜索结果' not in line:
|
||||||
|
break
|
||||||
|
|
||||||
|
if not in_results:
|
||||||
|
continue
|
||||||
|
|
||||||
|
# 匹配标题链接:link "标题文字" [ref=eXX]
|
||||||
|
# 需要过滤域名链接(如 "zhihu.com")和短链接
|
||||||
|
if 'link "' in line and '[ref=' in line:
|
||||||
|
match = re.search(r'link "([^"]+)" \[ref=(e\d+)\]', line)
|
||||||
|
if match:
|
||||||
|
title = match.group(1)
|
||||||
|
ref = match.group(2)
|
||||||
|
# 过滤短标题(域名链接如 "zhihu.com")
|
||||||
|
if len(title) > 20 and '.' not in title[:10]: # 不是域名格式
|
||||||
|
refs.append((title, ref))
|
||||||
|
|
||||||
|
# 获取每个结果的 URL
|
||||||
|
for title, ref in refs[:max_results]:
|
||||||
|
url = self._get_link_url(ref)
|
||||||
|
if url and 'bing.com/search' not in url: # 过滤搜索结果页本身的链接
|
||||||
|
results.append({
|
||||||
|
'title': title,
|
||||||
|
'url': url,
|
||||||
|
'snippet': '',
|
||||||
|
'source': 'bing'
|
||||||
|
})
|
||||||
|
|
||||||
|
return results
|
||||||
|
|
||||||
|
def _get_link_url(self, ref):
|
||||||
|
"""通过 agent-browser 获取链接的 URL"""
|
||||||
|
try:
|
||||||
|
stdout, stderr, code = self._run_browser('get', 'attr', f'@{ref}', 'href', '--json', '--timeout', '5000')
|
||||||
|
if code == 0 and stdout:
|
||||||
|
data = json.loads(stdout)
|
||||||
|
return data.get('data', {}).get('value', '')
|
||||||
|
except Exception as e:
|
||||||
|
print(f"获取 URL 失败 (ref={ref}): {e}")
|
||||||
|
return None
|
||||||
|
|
||||||
def fetch_url_content(self, url):
|
def fetch_url_content(self, url):
|
||||||
"""抓取网页内容"""
|
"""抓取网页内容"""
|
||||||
try:
|
try:
|
||||||
|
|||||||
Reference in New Issue
Block a user