searxng/searx/engines/google_news.py

51 lines
1.6 KiB
Python
Raw Normal View History

2014-03-04 12:11:04 +00:00
#!/usr/bin/env python
from urllib import urlencode
from json import loads
2014-03-15 19:20:41 +00:00
from datetime import datetime
2014-03-04 12:11:04 +00:00
categories = ['news']
url = 'https://ajax.googleapis.com/'
2014-03-04 13:20:29 +00:00
search_url = url + 'ajax/services/search/news?v=2.0&start={offset}&rsz=large&safe=off&filter=off&{query}&hl={language}' # noqa
2014-03-04 12:11:04 +00:00
paging = True
language_support = True
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
def response(resp):
results = []
search_res = loads(resp.text)
if not search_res.get('responseData', {}).get('results'):
return []
for result in search_res['responseData']['results']:
2014-03-14 08:55:04 +00:00
# S.149 (159), library.pdf
2014-03-15 19:20:41 +00:00
# datetime.strptime("Mon, 10 Mar 2014 16:26:15 -0700",
# "%a, %d %b %Y %H:%M:%S %z")
2014-03-14 08:55:04 +00:00
# publishedDate = parse(result['publishedDate'])
2014-03-15 19:20:41 +00:00
publishedDate = datetime.strptime(
str.join(' ', result['publishedDate'].split(None)[0:5]),
"%a, %d %b %Y %H:%M:%S")
#utc_offset = timedelta(result['publishedDate'].split(None)[5])
# local = utc + offset
2014-03-14 08:55:04 +00:00
#publishedDate = publishedDate + utc_offset
2014-03-04 12:11:04 +00:00
results.append({'url': result['unescapedUrl'],
'title': result['titleNoFormatting'],
2014-03-15 19:20:41 +00:00
'publishedDate': publishedDate,
2014-03-04 12:11:04 +00:00
'content': result['content']})
return results