git.iko.soy/searx
on master help sign in
Merge pull request #1563 from Nachtalb/ne/fix-google-image-search
629b36dAlexandre Flament
searx/engines/google_images.py+21−17
compare with master
unifiedsplit
@@ -11,7 +11,6 @@
11"""
11"""
12
12
13from datetime import date, timedelta
13from datetime import date, timedelta
14−from json import loads
15from lxml import html
14from lxml import html
16from searx.url_utils import urlencode, urlparse, parse_qs
15from searx.url_utils import urlencode, urlparse, parse_qs
17
16
@@ -39,7 +38,6 @@ time_range_dict = {'day': 'd',
39# do search-request
38# do search-request
40def request(query, params):
39def request(query, params):
41 search_options = {
40 search_options = {
42− 'ijn': params['pageno'] - 1,
43 'start': (params['pageno'] - 1) * number_of_results
41 'start': (params['pageno'] - 1) * number_of_results
44 }
42 }
45
43
@@ -53,7 +51,7 @@ def request(query, params):
53 search_options['tbs'] = time_range_custom_attr.format(start=start, end=end)
51 search_options['tbs'] = time_range_custom_attr.format(start=start, end=end)
54
52
55 if safesearch and params['safesearch']:
53 if safesearch and params['safesearch']:
56− search_options['safe'] = 'on'
54+ search_options['safe'] = 'active'
57
55
58 params['url'] = search_url.format(query=urlencode({'q': query}),
56 params['url'] = search_url.format(query=urlencode({'q': query}),
59 search_options=urlencode(search_options))
57 search_options=urlencode(search_options))
@@ -63,24 +61,30 @@ def request(query, params):
63
61
64# get response from search-request
62# get response from search-request
65def response(resp):
63def response(resp):
64+ dom = html.fromstring(resp.text)
65+
66 results = []
66 results = []
67+ for element in dom.xpath('//div[@id="search"] //td'):
68+ link = element.xpath('./a')[0]
67
69
68− dom = html.fromstring(resp.text)
70+ google_url = urlparse(link.xpath('.//@href')[0])
71+ query = parse_qs(google_url.query)
72+ source_url = next(iter(query.get('q', [])), None)
69
73
70− # parse results
74+ title_parts = element.xpath('./cite//following-sibling::*/text()')
71− for img in dom.xpath('//a'):
75+ title_parts.extend(element.xpath('./cite//following-sibling::text()')[:-1])
72− r = {
76+
73− 'title': u' '.join(img.xpath('.//div[class="rg_ilmbg"]//text()')),
77+ result = {
78+ 'title': ''.join(title_parts),
74 'content': '',
79 'content': '',
75 'template': 'images.html',
80 'template': 'images.html',
81+ 'url': source_url,
82+ 'img_src': source_url,
83+ 'thumbnail_src': next(iter(link.xpath('.//img //@src')), None)
76 }
84 }
77− url = urlparse(img.xpath('.//@href')[0])
85+
78− query = parse_qs(url.query)
79− r['url'] = query['imgrefurl'][0]
80− r['img_src'] = query['imgurl'][0]
81− r['thumbnail_src'] = r['img_src']
86+ if not source_url or not result['thumbnail_src']:
82− # append result
87+ continue
88+
83− results.append(r)
89+ results.append(result)
84−
85− # return results
86 return results
90 return results