Ponysearch/searx/engines/google_news.py

68 lines
1.7 KiB
Python
Raw Normal View History

"""
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
"""
2014-03-04 13:11:04 +01:00
from urllib import urlencode
from json import loads
2014-03-18 13:19:50 +01:00
from dateutil import parser
2014-03-04 13:11:04 +01:00
2014-09-01 15:10:05 +02:00
# search-url
2014-03-04 13:11:04 +01:00
categories = ['news']
2014-09-01 15:10:05 +02:00
paging = True
language_support = True
2014-03-04 13:11:04 +01:00
2014-09-01 15:10:05 +02:00
# engine dependent config
2014-03-04 13:11:04 +01:00
url = 'https://ajax.googleapis.com/'
2015-01-31 16:38:03 +01:00
search_url = url + 'ajax/services/search/news?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={lang}'
2014-03-04 13:11:04 +01:00
2014-09-01 15:10:05 +02:00
# do search-request
2014-03-04 13:11:04 +01:00
def request(query, params):
offset = (params['pageno'] - 1) * 8
2014-09-01 15:10:05 +02:00
2014-03-04 13:11:04 +01:00
language = 'en-US'
if params['language'] != 'all':
language = params['language'].replace('_', '-')
2014-09-01 15:10:05 +02:00
2014-03-04 13:11:04 +01:00
params['url'] = search_url.format(offset=offset,
query=urlencode({'q': query}),
2015-01-31 16:38:03 +01:00
lang=language)
2014-09-01 15:10:05 +02:00
2014-03-04 13:11:04 +01:00
return params
2014-09-01 15:10:05 +02:00
# get response from search-request
2014-03-04 13:11:04 +01:00
def response(resp):
results = []
2014-09-01 15:10:05 +02:00
2014-03-04 13:11:04 +01:00
search_res = loads(resp.text)
2014-09-01 15:10:05 +02:00
# return empty array if there are no results
2014-03-04 13:11:04 +01:00
if not search_res.get('responseData', {}).get('results'):
return []
2014-09-01 15:10:05 +02:00
# parse results
2014-03-04 13:11:04 +01:00
for result in search_res['responseData']['results']:
2014-09-01 15:10:05 +02:00
# parse publishedDate
2014-03-18 13:19:50 +01:00
publishedDate = parser.parse(result['publishedDate'])
2015-01-31 16:38:03 +01:00
if 'url' not in result:
continue
2014-03-14 09:55:04 +01:00
2014-09-01 15:10:05 +02:00
# append result
2014-03-04 13:11:04 +01:00
results.append({'url': result['unescapedUrl'],
'title': result['titleNoFormatting'],
2014-03-15 20:20:41 +01:00
'publishedDate': publishedDate,
2014-03-04 13:11:04 +01:00
'content': result['content']})
2014-09-01 15:10:05 +02:00
# return results
2014-03-04 13:11:04 +01:00
return results