mirror of
https://github.com/searxng/searxng.git
synced 2022-08-25 14:23:18 +00:00
ad7a6e6e10
When selecting other languages than 'en', bing-video did not handle the language correct and gave very bad results. Since User-Agent is normaly rotated in searx, the behavior of a !biv search was unpredictable and paging was broken. Signed-off-by: Markus Heiser <markus.heiser@darmarit.de>
99 lines
2.9 KiB
Python
99 lines
2.9 KiB
Python
"""
|
|
Bing (Videos)
|
|
|
|
@website https://www.bing.com/videos
|
|
@provide-api yes (http://datamarket.azure.com/dataset/bing/search)
|
|
|
|
@using-api no
|
|
@results HTML
|
|
@stable no
|
|
@parse url, title, content, thumbnail
|
|
"""
|
|
|
|
from json import loads
|
|
from lxml import html
|
|
from searx.url_utils import urlencode
|
|
from searx.utils import match_language
|
|
|
|
from searx.engines.bing import _fetch_supported_languages, supported_languages_url, language_aliases
|
|
|
|
categories = ['videos']
|
|
paging = True
|
|
safesearch = True
|
|
time_range_support = True
|
|
number_of_results = 28
|
|
language_support = True
|
|
|
|
base_url = 'https://www.bing.com/'
|
|
search_string = 'videos/search'\
|
|
'?{query}'\
|
|
'&count={count}'\
|
|
'&first={first}'\
|
|
'&scope=video'\
|
|
'&FORM=QBLH'
|
|
time_range_string = '&qft=+filterui:videoage-lt{interval}'
|
|
time_range_dict = {'day': '1440',
|
|
'week': '10080',
|
|
'month': '43200',
|
|
'year': '525600'}
|
|
|
|
# safesearch definitions
|
|
safesearch_types = {2: 'STRICT',
|
|
1: 'DEMOTE',
|
|
0: 'OFF'}
|
|
|
|
|
|
# do search-request
|
|
def request(query, params):
|
|
offset = ((params['pageno'] - 1) * number_of_results) + 1
|
|
|
|
search_path = search_string.format(
|
|
query=urlencode({'q': query}),
|
|
count=number_of_results,
|
|
first=offset)
|
|
|
|
# safesearch cookie
|
|
params['cookies']['SRCHHPGUSR'] = \
|
|
'ADLT=' + safesearch_types.get(params['safesearch'], 'DEMOTE')
|
|
|
|
# language cookie
|
|
language = match_language(params['language'], supported_languages, language_aliases).lower()
|
|
params['cookies']['_EDGE_S'] = 'mkt=' + language + '&F=1'
|
|
|
|
# query and paging
|
|
params['url'] = base_url + search_path
|
|
|
|
# time range
|
|
if params['time_range'] in time_range_dict:
|
|
params['url'] += time_range_string.format(interval=time_range_dict[params['time_range']])
|
|
|
|
# bing videos did not like "older" versions < 70.0.1 when selectin other
|
|
# languages then 'en' .. very strange ?!?!
|
|
params['headers']['User-Agent'] = 'Mozilla/5.0 (X11; Linux x86_64; rv:73.0.1) Gecko/20100101 Firefox/73.0.1'
|
|
|
|
return params
|
|
|
|
|
|
# get response from search-request
|
|
def response(resp):
|
|
results = []
|
|
|
|
dom = html.fromstring(resp.text)
|
|
|
|
for result in dom.xpath('//div[@class="dg_u"]'):
|
|
try:
|
|
metadata = loads(result.xpath('.//div[@class="vrhdata"]/@vrhm')[0])
|
|
info = ' - '.join(result.xpath('.//div[@class="mc_vtvc_meta_block"]//span/text()')).strip()
|
|
content = '{0} - {1}'.format(metadata['du'], info)
|
|
thumbnail = '{0}th?id={1}'.format(base_url, metadata['thid'])
|
|
results.append({'url': metadata['murl'],
|
|
'thumbnail': thumbnail,
|
|
'title': metadata.get('vt', ''),
|
|
'content': content,
|
|
'template': 'videos.html'})
|
|
|
|
except:
|
|
continue
|
|
|
|
return results
|