Ponysearch/searx/engines/bing.py

85 lines
2.2 KiB
Python
Raw Normal View History

2014-09-01 14:38:59 +02:00
## Bing (Web)
#
2014-09-01 14:38:59 +02:00
# @website https://www.bing.com
# @provide-api yes (http://datamarket.azure.com/dataset/bing/search),
# max. 5000 query/month
#
2014-09-01 14:38:59 +02:00
# @using-api no (because of query limit)
# @results HTML (using search portal)
# @stable no (HTML can change)
# @parse url, title, content
#
# @todo publishedDate
2013-10-24 23:52:57 +02:00
from urllib import urlencode
from cgi import escape
2014-02-05 20:24:31 +01:00
from lxml import html
2015-01-25 20:14:37 +01:00
from searx.engines.xpath import extract_text
2013-10-24 23:52:57 +02:00
2014-09-01 14:38:59 +02:00
# engine dependent config
categories = ['general']
2014-01-29 21:14:38 +01:00
paging = True
2014-01-31 04:35:23 +01:00
language_support = True
2014-01-29 21:14:38 +01:00
2014-09-01 14:38:59 +02:00
# search-url
base_url = 'https://www.bing.com/'
search_string = 'search?{query}&first={offset}'
2013-10-24 23:52:57 +02:00
2014-09-02 17:13:44 +02:00
2014-09-01 14:38:59 +02:00
# do search-request
2013-10-24 23:52:57 +02:00
def request(query, params):
2014-01-29 21:14:38 +01:00
offset = (params['pageno'] - 1) * 10 + 1
2014-09-01 14:38:59 +02:00
2014-01-31 04:35:23 +01:00
if params['language'] == 'all':
language = 'en-US'
else:
language = params['language'].replace('_', '-')
2014-09-01 14:38:59 +02:00
search_path = search_string.format(
2014-01-31 04:35:23 +01:00
query=urlencode({'q': query, 'setmkt': language}),
2014-01-29 21:14:38 +01:00
offset=offset)
2014-01-31 04:35:23 +01:00
params['cookies']['SRCHHPGUSR'] = \
'NEWWND=0&NRSLT=-1&SRCHLANG=' + language.split('-')[0]
2014-09-01 14:38:59 +02:00
2013-10-24 23:52:57 +02:00
params['url'] = base_url + search_path
return params
2014-09-01 14:38:59 +02:00
# get response from search-request
2013-10-24 23:52:57 +02:00
def response(resp):
results = []
2014-09-01 14:38:59 +02:00
2013-10-24 23:52:57 +02:00
dom = html.fromstring(resp.content)
2014-09-01 14:38:59 +02:00
# parse results
2013-10-24 23:52:57 +02:00
for result in dom.xpath('//div[@class="sa_cc"]'):
link = result.xpath('.//h3/a')[0]
url = link.attrib.get('href')
2015-01-25 20:14:37 +01:00
title = extract_text(link)
content = escape(extract_text(result.xpath('.//p')))
2013-10-25 01:37:48 +02:00
2014-09-01 14:38:59 +02:00
# append result
results.append({'url': url,
'title': title,
2014-09-01 14:38:59 +02:00
'content': content})
# return results if something is found
2013-10-25 01:37:48 +02:00
if results:
return results
2014-09-01 14:38:59 +02:00
# parse results again if nothing is found yet
2013-10-25 01:37:48 +02:00
for result in dom.xpath('//li[@class="b_algo"]'):
link = result.xpath('.//h2/a')[0]
url = link.attrib.get('href')
2015-01-25 20:14:37 +01:00
title = extract_text(link)
content = escape(extract_text(result.xpath('.//p')))
2014-09-01 14:38:59 +02:00
# append result
results.append({'url': url,
'title': title,
2014-09-01 14:38:59 +02:00
'content': content})
# return results
2013-10-24 23:52:57 +02:00
return results