searxng/searx/engines/startpage.py

33 lines
1006 B
Python
Raw Normal View History

2013-10-19 16:29:39 +00:00
from urllib import quote
from lxml import html
2013-10-24 21:43:39 +00:00
from urlparse import urlparse
2013-10-19 16:29:39 +00:00
from cgi import escape
2013-10-19 16:30:08 +00:00
base_url = 'https://startpage.com/'
2013-10-19 16:29:39 +00:00
search_url = base_url+'do/search'
def request(query, params):
global search_url
query = quote(query.replace(' ', '+'), safe='+')
params['url'] = search_url
params['method'] = 'POST'
params['data'] = {'query': query}
return params
def response(resp):
global base_url
results = []
2013-10-24 19:00:44 +00:00
dom = html.fromstring(resp.content)
2013-10-19 16:29:39 +00:00
for result in dom.xpath('//div[@class="result"]'):
link = result.xpath('.//h3/a')[0]
2013-10-24 21:43:39 +00:00
url = link.attrib.get('href')
2013-10-19 18:12:22 +00:00
parsed_url = urlparse(url)
# TODO better google link detection
2013-10-19 18:13:48 +00:00
if parsed_url.netloc.find('www.google.com') >= 0:
2013-10-19 18:12:22 +00:00
continue
2013-10-19 16:29:39 +00:00
title = ' '.join(link.xpath('.//text()'))
content = escape(' '.join(result.xpath('.//p[@class="desc"]//text()')))
results.append({'url': url, 'title': title, 'content': content})
return results