mirror of
https://github.com/searxng/searxng
synced 2024-01-01 18:24:07 +00:00
70 lines
1.6 KiB
Python
70 lines
1.6 KiB
Python
## DuckDuckGo (Web)
|
|
#
|
|
# @website https://duckduckgo.com/
|
|
# @provide-api yes (https://duckduckgo.com/api), but not all results from search-site
|
|
#
|
|
# @using-api no
|
|
# @results HTML (using search portal)
|
|
# @stable no (HTML can change)
|
|
# @parse url, title, content
|
|
#
|
|
# @todo rewrite to api
|
|
# @todo language support
|
|
|
|
from urllib import urlencode
|
|
from lxml.html import fromstring
|
|
from searx.utils import html_to_text
|
|
|
|
# engine dependent config
|
|
categories = ['general']
|
|
paging = True
|
|
locale = 'us-en'
|
|
|
|
# search-url
|
|
url = 'https://duckduckgo.com/html?{query}&s={offset}'
|
|
|
|
# specific xpath variables
|
|
result_xpath = '//div[@class="results_links results_links_deep web-result"]' # noqa
|
|
url_xpath = './/a[@class="large"]/@href'
|
|
title_xpath = './/a[@class="large"]//text()'
|
|
content_xpath = './/div[@class="snippet"]//text()'
|
|
|
|
|
|
# do search-request
|
|
def request(query, params):
|
|
offset = (params['pageno'] - 1) * 30
|
|
|
|
params['url'] = url.format(
|
|
query=urlencode({'q': query, 'l': locale}),
|
|
offset=offset)
|
|
|
|
return params
|
|
|
|
|
|
# get response from search-request
|
|
def response(resp):
|
|
results = []
|
|
|
|
doc = fromstring(resp.text)
|
|
|
|
# parse results
|
|
for r in doc.xpath(result_xpath):
|
|
try:
|
|
res_url = r.xpath(url_xpath)[-1]
|
|
except:
|
|
continue
|
|
|
|
if not res_url:
|
|
continue
|
|
|
|
title = html_to_text(''.join(r.xpath(title_xpath)))
|
|
content = html_to_text(''.join(r.xpath(content_xpath)))
|
|
|
|
# append result
|
|
results.append({'title': title,
|
|
'content': content,
|
|
'url': res_url})
|
|
|
|
# return results
|
|
return results
|