git.iko.soy/searx
on master help sign in
Merge branch 'master' into patch-1
9012ad4Alexandre Flament
searx/engines/duckduckgo_images.py+7−4
compare with master
unifiedsplit
@@ -35,9 +35,12 @@ site_url = 'https://duckduckgo.com/?{query}&iar=images&iax=1&ia=images'
35
35
36# run query in site to get vqd number needed for requesting images
36# run query in site to get vqd number needed for requesting images
37# TODO: find a way to get this number without an extra request (is it a hash of the query?)
37# TODO: find a way to get this number without an extra request (is it a hash of the query?)
38−def get_vqd(query):
38+def get_vqd(query, headers):
39− res = get(site_url.format(query=urlencode({'q': query})))
39+ query_url = site_url.format(query=urlencode({'q': query}))
40+ res = get(query_url, headers=headers)
40 content = res.text
41 content = res.text
42+ if content.find('vqd=\'') == -1:
43+ raise Exception('Request failed')
41 vqd = content[content.find('vqd=\'') + 5:]
44 vqd = content[content.find('vqd=\'') + 5:]
42 vqd = vqd[:vqd.find('\'')]
45 vqd = vqd[:vqd.find('\'')]
43 return vqd
46 return vqd
@@ -47,7 +50,7 @@ def get_vqd(query):
47def request(query, params):
50def request(query, params):
48 # to avoid running actual external requests when testing
51 # to avoid running actual external requests when testing
49 if 'is_test' not in params:
52 if 'is_test' not in params:
50− vqd = get_vqd(query)
53+ vqd = get_vqd(query, params['headers'])
51 else:
54 else:
52 vqd = '12345'
55 vqd = '12345'
53
56
@@ -74,7 +77,7 @@ def response(resp):
74 try:
77 try:
75 res_json = loads(content)
78 res_json = loads(content)
76 except:
79 except:
77− return []
80+ raise Exception('Cannot parse results')
78
81
79 # parse results
82 # parse results
80 for result in res_json['results']:
83 for result in res_json['results']:
searx/engines/google_images.py+21−17
@@ -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
tests/unit/engines/test_duckduckgo_images.py+1−1
@@ -41,7 +41,7 @@ class TestDuckduckgoImagesEngine(SearxTestCase):
41 self.assertRaises(AttributeError, duckduckgo_images.response, '[]')
41 self.assertRaises(AttributeError, duckduckgo_images.response, '[]')
42
42
43 response = mock.Mock(text='If this error persists, please let us know: ops@duckduckgo.com')
43 response = mock.Mock(text='If this error persists, please let us know: ops@duckduckgo.com')
44− self.assertEqual(duckduckgo_images.response(response), [])
44+ self.assertRaises(Exception, duckduckgo_images.response, response)
45
45
46 json = u"""
46 json = u"""
47 {
47 {