on master help sign in
Merge branch 'master' into nyaa
commitc3232b0e1a82315e5936fb3c0731548efd66a0b3 authormisnyo <misnyo@users.noreply.github.com> dateMon 4 Sep 2017, 17:48 (9 years ago) parent01330f76888010
5 files changed+77−84
searx/engines/generalfile.py−62 searx/engines/gigablast.py+6−2 searx/engines/google_news.py+2−2 searx/settings.yml+13−12 tests/unit/engines/test_google_news.py+56−6
searx/engines/generalfile.pydeleted−62
@@ -1,62 +0,0 @@
1−"""
2− General Files (Files)
3−
4− @website http://www.general-files.org
5− @provide-api no (nothing found)
6−
7− @using-api no (because nothing found)
8− @results HTML (using search portal)
9− @stable no (HTML can change)
10− @parse url, title, content
11−
12− @todo detect torrents?
13−"""
14−
15−from lxml import html
16−
17−# engine dependent config
18−categories = ['files']
19−paging = True
20−
21−# search-url
22−base_url = 'http://www.general-file.com'
23−search_url = base_url + '/files-{letter}/{query}/{pageno}'
24−
25−# specific xpath variables
26−result_xpath = '//table[@class="block-file"]'
27−title_xpath = './/h2/a//text()'
28−url_xpath = './/h2/a/@href'
29−content_xpath = './/p//text()'
30−
31−
32−# do search-request
33−def request(query, params):
34−
35− params['url'] = search_url.format(query=query,
36− letter=query[0],
37− pageno=params['pageno'])
38−
39− return params
40−
41−
42−# get response from search-request
43−def response(resp):
44− results = []
45−
46− dom = html.fromstring(resp.text)
47−
48− # parse results
49− for result in dom.xpath(result_xpath):
50− url = result.xpath(url_xpath)[0]
51−
52− # skip fast download links
53− if not url.startswith('/'):
54− continue
55−
56− # append result
57− results.append({'url': base_url + url,
58− 'title': ''.join(result.xpath(title_xpath)),
59− 'content': ''.join(result.xpath(content_xpath))})
60−
61− # return results
62− return results
searx/engines/gigablast.py+6−2
@@ -10,6 +10,7 @@
1010 @parse url, title, content
1111"""
1212
13+import random
1314from json import loads
1415from time import time
1516from lxml.html import fromstring
@@ -32,7 +33,8 @@ search_string = 'search?{query}'\
3233 '&qh=0'\
3334 '&qlang={lang}'\
3435 '&ff={safesearch}'\
35− '&rxikd={rxikd}' # random number - 9 digits
36+ '&rxieu={rxieu}'\
37+ '&rand={rxikd}' # current unix timestamp
3638
3739# specific xpath variables
3840results_xpath = '//response//result'
@@ -59,10 +61,12 @@ def request(query, params):
5961 else:
6062 safesearch = 0
6163
64+ # rxieu is some kind of hash from the search query, but accepts random atm
6265 search_path = search_string.format(query=urlencode({'q': query}),
6366 offset=offset,
6467 number_of_results=number_of_results,
65− rxikd=str(time())[:9],
68+ rxikd=int(time() * 1000),
69+ rxieu=random.randint(1000000000, 9999999999),
6670 lang=language,
6771 safesearch=safesearch)
6872
searx/engines/google_news.py+2−2
@@ -67,8 +67,8 @@ def response(resp):
6767 for result in dom.xpath('//div[@class="g"]|//div[@class="g _cy"]'):
6868 try:
6969 r = {
70− 'url': result.xpath('.//div[@class="_cnc"]//a/@href')[0],
71− 'title': ''.join(result.xpath('.//div[@class="_cnc"]//h3//text()')),
70+ 'url': result.xpath('.//a[@class="l _PMs"]')[0].attrib.get("href"),
71+ 'title': ''.join(result.xpath('.//a[@class="l _PMs"]//text()')),
7272 'content': ''.join(result.xpath('.//div[@class="st"]//text()')),
7373 }
7474 except:
searx/settings.yml+13−12
@@ -242,15 +242,16 @@ engines:
242242 disabled: True
243243
244244 - name : gitlab
245− engine : xpath
245+ engine : json_engine
246246 paging : True
247− search_url : https://gitlab.com/search?page={pageno}&search={query}
248− url_xpath : //li[@class="project-row"]//a[@class="project"]/@href
249− title_xpath : //li[@class="project-row"]//span[contains(@class, "project-full-name")]
250− content_xpath : //li[@class="project-row"]//div[@class="description"]/p
247+ search_url : https://gitlab.com/api/v4/projects?search={query}&page={pageno}
248+ url_query : web_url
249+ title_query : name_with_namespace
250+ content_query : description
251+ page_size : 20
251252 categories : it
252253 shortcut : gl
253− timeout : 5.0
254+ timeout : 10.0
254255 disabled : True
255256
256257 - name : github
@@ -321,9 +322,9 @@ engines:
321322 engine : xpath
322323 paging : True
323324 search_url : https://geektimes.ru/search/page{pageno}/?q={query}
324− url_xpath : //div[@class="search_results"]//a[@class="post__title_link"]/@href
325− title_xpath : //div[@class="search_results"]//a[@class="post__title_link"]
326− content_xpath : //div[@class="search_results"]//div[contains(@class, "content")]
325+ url_xpath : //article[contains(@class, "post")]//a[@class="post__title_link"]/@href
326+ title_xpath : //article[contains(@class, "post")]//a[@class="post__title_link"]
327+ content_xpath : //article[contains(@class, "post")]//div[contains(@class, "post__text")]
327328 categories : it
328329 timeout : 4.0
329330 disabled : True
@@ -333,9 +334,9 @@ engines:
333334 engine : xpath
334335 paging : True
335336 search_url : https://habrahabr.ru/search/page{pageno}/?q={query}
336− url_xpath : //div[@class="search_results"]//a[contains(@class, "post__title_link")]/@href
337− title_xpath : //div[@class="search_results"]//a[contains(@class, "post__title_link")]
338− content_xpath : //div[@class="search_results"]//div[contains(@class, "content")]
337+ url_xpath : //article[contains(@class, "post")]//a[@class="post__title_link"]/@href
338+ title_xpath : //article[contains(@class, "post")]//a[@class="post__title_link"]
339+ content_xpath : //article[contains(@class, "post")]//div[contains(@class, "post__text")]
339340 categories : it
340341 timeout : 4.0
341342 disabled : True
tests/unit/engines/test_google_news.py+56−6
@@ -37,14 +37,64 @@ class TestGoogleNewsEngine(SearxTestCase):
3737 self.assertEqual(google_news.response(response), [])
3838
3939 html = u"""
40−<div class="g">
41−<div class="ts _V6c _Zmc _XO _knc _d7c"><a class="top _vQb _mnc" href="http://this.is.the.url" onmousedown="return rwt(this,'','','','5','AFQjCNGixEtJGC3qTB9pYFLXlRj8XXwdiA','','0ahUKEwiG7O_M5-rQAhWDtRoKHd0RD5QQvIgBCCwwBA','','',event)"><img class="th _lub" id="news-thumbnail-image-52779299683347" src="data:image/jpeg;base64,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" alt="A(z) south témájának képe a következőből: CBC.ca" data-deferred="1" onload="google.aft&amp;&amp;google.aft(this)"></a><div class="_cnc"><h3 class="r _U6c"><a class="l _HId" href="http://this.is.the.url" onmousedown="return rwt(this,'','','','5','AFQjCNGixEtJGC3qTB9pYFLXlRj8XXwdiA','','0ahUKEwiG7O_M5-rQAhWDtRoKHd0RD5QQqQIILSgAMAQ','','',event)">Meet Thuli Madonsela — <em>South</em> Africa's conscience</a></h3><div class="slp"><span class="_tQb _IId">CBC.ca</span><span class="_v5">-</span><span class="f nsa _uQb">9 órával ezelőtt</span></div><div class="st"><em>South</em> African Public Protector</div></div><div class="_Xmc card-section"><a class="_sQb" href="http://www.news24.com/Columnists/Mpumelelo_Mkhabela/who-really-governs-south-africa-20161209" onmousedown="return rwt(this,'','','','5','AFQjCNHhc2MnYSZ5T4COqInzvgoju5k5bA','','0ahUKEwiG7O_M5-rQAhWDtRoKHd0RD5QQuogBCC4oATAE','','',event)">Who really governs <em>South</em> Africa?</a><br><span class="_Wmc _GId">Vélemény</span><span class="_v5">-</span><span class="_tQb _IId">News24</span><span class="_v5">-</span><span class="f nsa _uQb">2016. dec. 8.</span></div><div class="_Vmc"></div></div>
40+<h2 class="hd">Search Results</h2>
41+<div data-async-context="query:searx" id="ires">
42+ <div eid="oC2oWcGXCafR6ASkwoCwDA" id="rso">
43+ <div class="_NId">
44+ <!--m-->
45+ <div class="g _cy">
46+ <div class="ts _JGs _JHs _tJs _KGs _jHs">
47+ <div class="_hJs">
48+ <h3 class="r _gJs">
49+ <a class="l _PMs" href="https://example.com/" onmousedown="return rwt(this,'','','','11','AFQjCNEyehpzD5cJK1KUfXBx9RmsbqqG9g','','0ahUKEwjB58OR54HWAhWnKJoKHSQhAMY4ChCpAggiKAAwAA','','',event)">Example title</a>
50+ </h3>
51+ <div class="slp">
52+ <span class="_OHs _PHs">
53+ Mac &amp; i</span>
54+ <span class="_QGs">
55+ -</span>
56+ <span class="f nsa _QHs">
57+ Mar 21, 2016</span>
58+ </div>
59+ <div class="st">Example description</div>
60+ </div>
61+ </div>
62+ </div>
63+ <div class="g _cy">
64+ <div class="ts _JGs _JHs _oGs _KGs _jHs">
65+ <a class="top _xGs _SHs" href="https://example2.com/" onmousedown="return rwt(this,'','','','12','AFQjCNHObfH7sYmLWI1SC-YhWXKZFRzRjw','','0ahUKEwjB58OR54HWAhWnKJoKHSQhAMY4ChC8iAEIJDAB','','',event)">
66+ <img class="th _RGs" src="https://example2.com/image.jpg" alt="Story image for searx from Golem.de" onload="typeof google==='object'&&google.aft&&google.aft(this)">
67+ </a>
68+ <div class="_hJs">
69+ <h3 class="r _gJs">
70+ <a class="l _PMs" href="https://example2.com/" onmousedown="return rwt(this,'','','','12','AFQjCNHObfH7sYmLWI1SC-YhWXKZFRzRjw','','0ahUKEwjB58OR54HWAhWnKJoKHSQhAMY4ChCpAgglKAAwAQ','','',event)">Example title 2</a>
71+ </h3>
72+ <div class="slp">
73+ <span class="_OHs _PHs">
74+ Golem.de</span>
75+ <span class="_QGs">
76+ -</span>
77+ <span class="f nsa _QHs">
78+ Oct 4, 2016</span>
79+ </div>
80+ <div class="st">Example description 2</div>
81+ </div>
82+ </div>
83+ </div>
84+ </div>
85+ </div>
4286</div>
87+
88+
4389 """ # noqa
4490 response = mock.Mock(text=html)
4591 results = google_news.response(response)
4692 self.assertEqual(type(results), list)
47− self.assertEqual(len(results), 1)
48− self.assertEqual(results[0]['title'], u'Meet Thuli Madonsela \u2014 South Africa\'s conscience')
49− self.assertEqual(results[0]['url'], 'http://this.is.the.url')
50− self.assertEqual(results[0]['content'], 'South African Public Protector')
93+ self.assertEqual(len(results), 2)
94+ self.assertEqual(results[0]['title'], u'Example title')
95+ self.assertEqual(results[0]['url'], 'https://example.com/')
96+ self.assertEqual(results[0]['content'], 'Example description')
97+ self.assertEqual(results[1]['title'], u'Example title 2')
98+ self.assertEqual(results[1]['url'], 'https://example2.com/')
99+ self.assertEqual(results[1]['content'], 'Example description 2')
100+ self.assertEqual(results[1]['img_src'], 'https://example2.com/image.jpg')