mirror of
https://github.com/searxng/searxng.git
synced 2024-12-23 17:50:29 +00:00
87 lines
2.1 KiB
Python
87 lines
2.1 KiB
Python
"""
|
|
Google (Images)
|
|
|
|
@website https://www.google.com
|
|
@provide-api yes (https://developers.google.com/custom-search/)
|
|
|
|
@using-api no
|
|
@results HTML chunks with JSON inside
|
|
@stable no
|
|
@parse url, title, img_src
|
|
"""
|
|
|
|
from urllib import urlencode
|
|
from json import loads
|
|
from lxml import html
|
|
|
|
# engine dependent config
|
|
categories = ['images']
|
|
paging = True
|
|
safesearch = True
|
|
time_range_support = True
|
|
number_of_results = 100
|
|
|
|
search_url = 'https://www.google.com/search'\
|
|
'?{query}'\
|
|
'&asearch=ichunk'\
|
|
'&async=_id:rg_s,_pms:s'\
|
|
'&tbm=isch'\
|
|
'&yv=2'\
|
|
'&{search_options}'
|
|
time_range_attr = "qdr:{range}"
|
|
time_range_dict = {'day': 'd',
|
|
'week': 'w',
|
|
'month': 'm'}
|
|
|
|
|
|
# do search-request
|
|
def request(query, params):
|
|
|
|
search_options = {
|
|
'ijn': params['pageno'] - 1,
|
|
'start': (params['pageno'] - 1) * number_of_results
|
|
}
|
|
|
|
if params['time_range'] in time_range_dict:
|
|
search_options['tbs'] = time_range_attr.format(range=time_range_dict[params['time_range']])
|
|
|
|
if safesearch and params['safesearch']:
|
|
search_options['safe'] = 'on'
|
|
|
|
params['url'] = search_url.format(query=urlencode({'q': query}),
|
|
search_options=urlencode(search_options))
|
|
|
|
return params
|
|
|
|
|
|
# get response from search-request
|
|
def response(resp):
|
|
results = []
|
|
|
|
g_result = loads(resp.text)
|
|
|
|
dom = html.fromstring(g_result[1][1])
|
|
|
|
# parse results
|
|
for result in dom.xpath('//div[@data-ved]'):
|
|
|
|
try:
|
|
metadata = loads(''.join(result.xpath('./div[@class="rg_meta"]/text()')))
|
|
except:
|
|
continue
|
|
|
|
thumbnail_src = metadata['tu']
|
|
|
|
# http to https
|
|
thumbnail_src = thumbnail_src.replace("http://", "https://")
|
|
|
|
# append result
|
|
results.append({'url': metadata['ru'],
|
|
'title': metadata['pt'],
|
|
'content': metadata['s'],
|
|
'thumbnail_src': thumbnail_src,
|
|
'img_src': metadata['ou'],
|
|
'template': 'images.html'})
|
|
|
|
# return results
|
|
return results
|