mirror of
https://github.com/searxng/searxng
synced 2024-11-09 01:10:26 +00:00
6bf9c398a7
If no language is specified, bing returns results with multiple languages for one query which isn't really useful. Setting english as default insted if nothing.
100 lines
2.6 KiB
Python
100 lines
2.6 KiB
Python
"""
|
|
Bing (Web)
|
|
|
|
@website https://www.bing.com
|
|
@provide-api yes (http://datamarket.azure.com/dataset/bing/search),
|
|
max. 5000 query/month
|
|
|
|
@using-api no (because of query limit)
|
|
@results HTML (using search portal)
|
|
@stable no (HTML can change)
|
|
@parse url, title, content
|
|
|
|
@todo publishedDate
|
|
"""
|
|
|
|
from urllib import urlencode
|
|
from lxml import html
|
|
from searx.engines.xpath import extract_text
|
|
|
|
# engine dependent config
|
|
categories = ['general']
|
|
paging = True
|
|
language_support = True
|
|
supported_languages_url = 'https://www.bing.com/account/general'
|
|
|
|
# search-url
|
|
base_url = 'https://www.bing.com/'
|
|
search_string = 'search?{query}&first={offset}'
|
|
|
|
|
|
# do search-request
|
|
def request(query, params):
|
|
offset = (params['pageno'] - 1) * 10 + 1
|
|
|
|
if params['language'] != 'all':
|
|
lang = params['language'].split('-')[0].upper()
|
|
else:
|
|
lang = 'EN'
|
|
|
|
query = u'language:{} {}'.format(lang, query.decode('utf-8')).encode('utf-8')
|
|
|
|
search_path = search_string.format(
|
|
query=urlencode({'q': query}),
|
|
offset=offset)
|
|
|
|
params['url'] = base_url + search_path
|
|
return params
|
|
|
|
|
|
# get response from search-request
|
|
def response(resp):
|
|
results = []
|
|
|
|
dom = html.fromstring(resp.text)
|
|
|
|
try:
|
|
results.append({'number_of_results': int(dom.xpath('//span[@class="sb_count"]/text()')[0]
|
|
.split()[0].replace(',', ''))})
|
|
except:
|
|
pass
|
|
|
|
# parse results
|
|
for result in dom.xpath('//div[@class="sa_cc"]'):
|
|
link = result.xpath('.//h3/a')[0]
|
|
url = link.attrib.get('href')
|
|
title = extract_text(link)
|
|
content = extract_text(result.xpath('.//p'))
|
|
|
|
# append result
|
|
results.append({'url': url,
|
|
'title': title,
|
|
'content': content})
|
|
|
|
# parse results again if nothing is found yet
|
|
for result in dom.xpath('//li[@class="b_algo"]'):
|
|
link = result.xpath('.//h2/a')[0]
|
|
url = link.attrib.get('href')
|
|
title = extract_text(link)
|
|
content = extract_text(result.xpath('.//p'))
|
|
|
|
# append result
|
|
results.append({'url': url,
|
|
'title': title,
|
|
'content': content})
|
|
|
|
# return results
|
|
return results
|
|
|
|
|
|
# get supported languages from their site
|
|
def _fetch_supported_languages(resp):
|
|
supported_languages = []
|
|
dom = html.fromstring(resp.text)
|
|
options = dom.xpath('//div[@id="limit-languages"]//input')
|
|
for option in options:
|
|
code = option.xpath('./@id')[0].replace('_', '-')
|
|
supported_languages.append(code)
|
|
|
|
return supported_languages
|