[mod] ddg engine mods

This commit is contained in:
Adam Tauber 2014-03-21 16:33:17 +01:00
parent ce08abe223
commit 3854703d95
1 changed files with 45 additions and 13 deletions

View File

@ -1,29 +1,61 @@
from json import loads
from urllib import urlencode from urllib import urlencode
from lxml.html import fromstring
from searx.utils import html_to_text from searx.utils import html_to_text
url = 'https://duckduckgo.com/' url = 'https://duckduckgo.com/html?{query}&s={offset}'
search_url = url + 'd.js?{query}&p=1&s={offset}'
locale = 'us-en' locale = 'us-en'
paging = True
def request(query, params): def request(query, params):
offset = (params['pageno'] - 1) * 30 offset = (params['pageno'] - 1) * 30
q = urlencode({'q': query, q = urlencode({'q': query,
'l': locale}) 'l': locale})
params['url'] = search_url.format(query=q, offset=offset) params['url'] = url.format(query=q, offset=offset)
return params return params
def response(resp): def response(resp):
result_xpath = '//div[@class="results_links results_links_deep web-result"]'
url_xpath = './/a[@class="large"]/@href'
title_xpath = './/a[@class="large"]//text()'
content_xpath = './/div[@class="snippet"]//text()'
results = [] results = []
search_res = loads(resp.text[resp.text.find('[{'):-2])[:-1]
for r in search_res: doc = fromstring(resp.text)
if not r.get('t'):
for r in doc.xpath(result_xpath):
res_url = r.xpath(url_xpath)[-1]
if not res_url:
continue continue
results.append({'title': r['t'], title = html_to_text(''.join(r.xpath(title_xpath)))
'content': html_to_text(r['a']), content = html_to_text(''.join(r.xpath(content_xpath)))
'url': r['u']}) results.append({'title': title,
'content': content,
'url': res_url})
return results return results
#from json import loads
#search_url = url + 'd.js?{query}&p=1&s={offset}'
#
#paging = True
#
#
#def request(query, params):
# offset = (params['pageno'] - 1) * 30
# q = urlencode({'q': query,
# 'l': locale})
# params['url'] = search_url.format(query=q, offset=offset)
# return params
#
#
#def response(resp):
# results = []
# search_res = loads(resp.text[resp.text.find('[{'):-2])[:-1]
# for r in search_res:
# if not r.get('t'):
# continue
# results.append({'title': r['t'],
# 'content': html_to_text(r['a']),
# 'url': r['u']})
# return results