Use string formatter to create source and img_format labels (#1566)

google_images :  use JSON embedded in HTML (engine expected pure JSON)
This commit is contained in:
Frank de Lange 2019-05-28 05:33:31 +02:00 committed by Alexandre Flament
parent a255651671
commit 4b7332286a

View File

@ -11,9 +11,9 @@
""" """
from datetime import date, timedelta from datetime import date, timedelta
from json import loads
from lxml import html from lxml import html
from searx.url_utils import urlencode, urlparse, parse_qs from searx.url_utils import urlencode
# engine dependent config # engine dependent config
categories = ['images'] categories = ['images']
@ -25,8 +25,7 @@ number_of_results = 100
search_url = 'https://www.google.com/search'\ search_url = 'https://www.google.com/search'\
'?{query}'\ '?{query}'\
'&tbm=isch'\ '&tbm=isch'\
'&gbv=1'\ '&yv=2'\
'&sa=G'\
'&{search_options}' '&{search_options}'
time_range_attr = "qdr:{range}" time_range_attr = "qdr:{range}"
time_range_custom_attr = "cdr:1,cd_min:{start},cd_max{end}" time_range_custom_attr = "cdr:1,cd_min:{start},cd_max{end}"
@ -38,6 +37,7 @@ time_range_dict = {'day': 'd',
# do search-request # do search-request
def request(query, params): def request(query, params):
search_options = { search_options = {
'ijn': params['pageno'] - 1,
'start': (params['pageno'] - 1) * number_of_results 'start': (params['pageno'] - 1) * number_of_results
} }
@ -51,7 +51,7 @@ def request(query, params):
search_options['tbs'] = time_range_custom_attr.format(start=start, end=end) search_options['tbs'] = time_range_custom_attr.format(start=start, end=end)
if safesearch and params['safesearch']: if safesearch and params['safesearch']:
search_options['safe'] = 'active' search_options['safe'] = 'on'
params['url'] = search_url.format(query=urlencode({'q': query}), params['url'] = search_url.format(query=urlencode({'q': query}),
search_options=urlencode(search_options)) search_options=urlencode(search_options))
@ -61,30 +61,27 @@ def request(query, params):
# get response from search-request # get response from search-request
def response(resp): def response(resp):
results = []
dom = html.fromstring(resp.text) dom = html.fromstring(resp.text)
results = [] # parse results
for element in dom.xpath('//div[@id="search"] //td'): for result in dom.xpath('//div[contains(@class, "rg_meta")]/text()'):
link = element.xpath('./a')[0]
google_url = urlparse(link.xpath('.//@href')[0]) try:
query = parse_qs(google_url.query) metadata = loads(result)
source_url = next(iter(query.get('q', [])), None) img_format = "{0} {1}x{2}".format(metadata['ity'], str(metadata['ow']), str(metadata['oh']))
source = "{0} ({1})".format(metadata['st'], metadata['isu'])
results.append({'url': metadata['ru'],
'title': metadata['pt'],
'content': metadata['s'],
'source': source,
'img_format': img_format,
'thumbnail_src': metadata['tu'],
'img_src': metadata['ou'],
'template': 'images.html'})
title_parts = element.xpath('./cite//following-sibling::*/text()') except:
title_parts.extend(element.xpath('./cite//following-sibling::text()')[:-1])
result = {
'title': ''.join(title_parts),
'content': '',
'template': 'images.html',
'url': source_url,
'img_src': source_url,
'thumbnail_src': next(iter(link.xpath('.//img //@src')), None)
}
if not source_url or not result['thumbnail_src']:
continue continue
results.append(result)
return results return results