1
0
Fork 0
mirror of https://github.com/YunoHost-Apps/searx_ynh.git synced 2024-09-03 20:16:30 +02:00
searx_ynh/sources/searx/engines/google_news.py
2014-12-01 12:26:38 +01:00

62 lines
1.7 KiB
Python

## Google (News)
#
# @website https://www.google.com
# @provide-api yes (https://developers.google.com/web-search/docs/), deprecated!
#
# @using-api yes
# @results JSON
# @stable yes (but deprecated)
# @parse url, title, content, publishedDate
from urllib import urlencode
from json import loads
from dateutil import parser
# search-url
categories = ['news']
paging = True
language_support = True
# engine dependent config
url = 'https://ajax.googleapis.com/'
search_url = url + 'ajax/services/search/news?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
# do search-request
def request(query, params):
offset = (params['pageno'] - 1) * 8
language = 'en-US'
if params['language'] != 'all':
language = params['language'].replace('_', '-')
params['url'] = search_url.format(offset=offset,
query=urlencode({'q': query}),
language=language)
return params
# get response from search-request
def response(resp):
results = []
search_res = loads(resp.text)
# return empty array if there are no results
if not search_res.get('responseData', {}).get('results'):
return []
# parse results
for result in search_res['responseData']['results']:
# parse publishedDate
publishedDate = parser.parse(result['publishedDate'])
# append result
results.append({'url': result['unescapedUrl'],
'title': result['titleNoFormatting'],
'publishedDate': publishedDate,
'content': result['content']})
# return results
return results