on master help sign in
Merge pull request #523 from a01200356/master

[fix] duckduckgo's xpaths changed

commit24ea39d04603cb28436af77450c233c0d53d63ae authorAdam Tauber <asciimoo@gmail.com> dateTue 22 Mar 2016, 11:12 (11 years ago) parentdbf4f83a5238ce
3 files changed+27−47
.travis.yml+1−1 searx/engines/duckduckgo.py+4−4 tests/unit/engines/test_duckduckgo.py+22−42
.travis.yml+1−1
@@ -10,7 +10,7 @@ python:
1010before_install:
1111 - "export DISPLAY=:99.0"
1212 - "sh -e /etc/init.d/xvfb start"
13− - npm install -g less grunt-cli
13+ - npm install less grunt-cli
1414 - ( cd searx/static/themes/oscar;npm install; cd - )
1515install:
1616 - ./manage.sh update_dev_packages
searx/engines/duckduckgo.py+4−4
@@ -28,10 +28,10 @@ language_support = True
2828url = 'https://duckduckgo.com/html?{query}&s={offset}'
2929
3030# specific xpath variables
31−result_xpath = '//div[@class="results_links results_links_deep web-result"]' # noqa
32−url_xpath = './/a[@class="large"]/@href'
33−title_xpath = './/a[@class="large"]'
34−content_xpath = './/div[@class="snippet"]'
31+result_xpath = '//div[@class="result results_links results_links_deep web-result "]' # noqa
32+url_xpath = './/a[@class="result__a"]/@href'
33+title_xpath = './/a[@class="result__a"]'
34+content_xpath = './/a[@class="result__snippet"]'
3535
3636
3737# do search-request
tests/unit/engines/test_duckduckgo.py+22−42
@@ -32,55 +32,32 @@ class TestDuckduckgoEngine(SearxTestCase):
3232 self.assertEqual(duckduckgo.response(response), [])
3333
3434 html = u"""
35− <div class="results_links results_links_deep web-result">
36− <div class="icon_fav" style="display: block;">
37− <a rel="nofollow" href="https://www.test.com/">
38− <img width="16" height="16" alt=""
39− src="/i/www.test.com.ico" style="visibility: visible;" name="i15" />
40− </a>
41− </div>
42− <div class="links_main links_deep"> <!-- This is the visible part -->
43− <a rel="nofollow" class="large" href="http://this.should.be.the.link/ű">
44− This <b>is</b> <b>the</b> title
45− </a>
46− <div class="snippet"><b>This</b> should be the content.</div>
47− <div class="url">
48− http://this.should.be.the.link/
35+ <div class="result results_links results_links_deep web-result result--no-result">
36+ <div class="links_main links_deep result__body">
37+ <h2 class="result__title">
38+ </h2>
39+ <div class="no-results">No results</div>
40+ <div class="result__extras">
4941 </div>
5042 </div>
5143 </div>
5244 """
5345 response = mock.Mock(text=html)
5446 results = duckduckgo.response(response)
55− self.assertEqual(type(results), list)
56− self.assertEqual(len(results), 1)
57− self.assertEqual(results[0]['title'], 'This is the title')
58− self.assertEqual(results[0]['url'], u'http://this.should.be.the.link/ű')
59− self.assertEqual(results[0]['content'], 'This should be the content.')
47+ self.assertEqual(duckduckgo.response(response), [])
6048
61− html = """
62− <div class="results_links results_links_deep web-result">
63− <div class="icon_fav" style="display: block;">
64− </div>
65− <div class="links_main links_deep"> <!-- This is the visible part -->
66− <div class="snippet"><b>This</b> should be the content.</div>
67− <div class="url">
68− http://this.should.be.the.link/
69− </div>
70− </div>
71− </div>
72− <div class="results_links results_links_deep web-result">
73− <div class="icon_fav" style="display: block;">
74− <img width="16" height="16" alt=""
75− src="/i/www.test.com.ico" style="visibility: visible;" name="i15" />
76− </div>
77− <div class="links_main links_deep"> <!-- This is the visible part -->
78− <a rel="nofollow" class="large" href="">
79− This <b>is</b> <b>the</b> title
49+ html = u"""
50+ <div class="result results_links results_links_deep web-result ">
51+ <div class="links_main links_deep result__body">
52+ <h2 class="result__title">
53+ <a rel="nofollow" class="result__a" href="http://this.should.be.the.link/ű">
54+ This <b>is</b> <b>the</b> title
55+ </a>
56+ </h2>
57+ <a class="result__snippet" href="http://this.should.be.the.link/ű">
58+ <b>This</b> should be the content.
8059 </a>
81− <div class="snippet"><b>This</b> should be the content.</div>
82− <div class="url">
83− http://this.should.be.the.link/
60+ <div class="result__extras">
8461 </div>
8562 </div>
8663 </div>
@@ -88,4 +65,7 @@ class TestDuckduckgoEngine(SearxTestCase):
8865 response = mock.Mock(text=html)
8966 results = duckduckgo.response(response)
9067 self.assertEqual(type(results), list)
91− self.assertEqual(len(results), 0)
68+ self.assertEqual(len(results), 1)
69+ self.assertEqual(results[0]['title'], 'This is the title')
70+ self.assertEqual(results[0]['url'], u'http://this.should.be.the.link/ű')
71+ self.assertEqual(results[0]['content'], 'This should be the content.')