on master help sign in
Add support for dokuwiki engine
commitcf09b500f35fd1bca3fc9cc853bd7ea932220e4e authorGuilhem Bonnefille <guilhem.bonnefille@gmail.com> dateSun 3 Apr 2016, 22:03 (11 years ago) parent5cbe4c5
2 files changed+169
searx/engines/doku.py+83 tests/unit/engines/test_doku.py+86
searx/engines/doku.pynew file+83
@@ -0,0 +1,83 @@
1+# Doku Wiki
2+#
3+# @website https://www.dokuwiki.org/
4+# @provide-api yes
5+# (https://www.dokuwiki.org/devel:xmlrpc)
6+#
7+# @using-api no
8+# @results HTML
9+# @stable yes
10+# @parse (general) url, title, content
11+
12+from urllib import urlencode
13+from lxml.html import fromstring
14+from searx.engines.xpath import extract_text
15+
16+# engine dependent config
17+categories = ['general'] # TODO , 'images', 'music', 'videos', 'files'
18+paging = False
19+language_support = False
20+number_of_results = 5
21+
22+# search-url
23+# Doku is OpenSearch compatible
24+base_url = 'http://localhost:8090'
25+search_url = '/?do=search'\
26+ '&id={query}'
27+# TODO '&startRecord={offset}'\
28+# TODO '&maximumRecords={limit}'\
29+
30+# do search-request
31+def request(query, params):
32+
33+ params['url'] = base_url +\
34+ search_url.format(query=urlencode({'query': query}))
35+
36+ return params
37+
38+
39+# get response from search-request
40+def response(resp):
41+ results = []
42+
43+ doc = fromstring(resp.text)
44+
45+ # parse results
46+ # Quickhits
47+ for r in doc.xpath('//div[@class="search_quickresult"]/ul/li'):
48+ try:
49+ res_url = r.xpath('.//a[@class="wikilink1"]/@href')[-1]
50+ except:
51+ continue
52+
53+ if not res_url:
54+ continue
55+
56+ title = extract_text(r.xpath('.//a[@class="wikilink1"]/@title'))
57+
58+ # append result
59+ results.append({'title': title,
60+ 'content': "",
61+ 'url': base_url + res_url})
62+
63+ # Search results
64+ for r in doc.xpath('//dl[@class="search_results"]/*'):
65+ try:
66+ if r.tag == "dt":
67+ res_url = r.xpath('.//a[@class="wikilink1"]/@href')[-1]
68+ title = extract_text(r.xpath('.//a[@class="wikilink1"]/@title'))
69+ elif r.tag == "dd":
70+ content = extract_text(r.xpath('.'))
71+
72+ # append result
73+ results.append({'title': title,
74+ 'content': content,
75+ 'url': base_url + res_url})
76+ except:
77+ continue
78+
79+ if not res_url:
80+ continue
81+
82+ # return results
83+ return results
tests/unit/engines/test_doku.pynew file+86
@@ -0,0 +1,86 @@
1+# -*- coding: utf-8 -*-
2+from collections import defaultdict
3+import mock
4+from searx.engines import doku
5+from searx.testing import SearxTestCase
6+
7+
8+class TestDokuEngine(SearxTestCase):
9+
10+ def test_request(self):
11+ query = 'test_query'
12+ dicto = defaultdict(dict)
13+ params = doku.request(query, dicto)
14+ self.assertIn('url', params)
15+ self.assertIn(query, params['url'])
16+
17+ def test_response(self):
18+ self.assertRaises(AttributeError, doku.response, None)
19+ self.assertRaises(AttributeError, doku.response, [])
20+ self.assertRaises(AttributeError, doku.response, '')
21+ self.assertRaises(AttributeError, doku.response, '[]')
22+
23+ response = mock.Mock(text='<html></html>')
24+ self.assertEqual(doku.response(response), [])
25+
26+ html = u"""
27+ <div class="search_quickresult">
28+ <h3>Pages trouvées :</h3>
29+ <ul class="search_quickhits">
30+ <li> <a href="/xfconf-query" class="wikilink1" title="xfconf-query">xfconf-query</a></li>
31+ </ul>
32+ <div class="clearer"></div>
33+ </div>
34+ """
35+ response = mock.Mock(text=html)
36+ results = doku.response(response)
37+ self.assertEqual(doku.response(response), [{'content': '', 'title': 'xfconf-query', 'url': 'http://localhost:8090/xfconf-query'}])
38+
39+ html = u"""
40+ <dl class="search_results">
41+ <dt><a href="/xvnc?s[]=query" class="wikilink1" title="xvnc">xvnc</a>: 40 Occurrences trouvées</dt>
42+ <dd>er = /usr/bin/Xvnc
43+ server_args = -inetd -<strong class="search_hit">query</strong> localhost -once -geometry 640x480 -depth 8 -Secur... er = /usr/bin/Xvnc
44+ server_args = -inetd -<strong class="search_hit">query</strong> localhost -once -geometry 800x600 -depth 8 -Secur... er = /usr/bin/Xvnc
45+ server_args = -inetd -<strong class="search_hit">query</strong> localhost -once -geometry 1024x768 -depth 8 -Secu... er = /usr/bin/Xvnc
46+ server_args = -inetd -<strong class="search_hit">query</strong> localhost -once -geometry 1280x1024 -depth 8 -Sec</dd>
47+ <dt><a href="/postfix_mysql_tls_sasl_1404?s[]=query" class="wikilink1" title="postfix_mysql_tls_sasl_1404">postfix_mysql_tls_sasl_1404</a>: 14 Occurrences trouvées</dt>
48+ <dd>tdepasse
49+ hosts = 127.0.0.1
50+ dbname = postfix
51+ <strong class="search_hit">query</strong> = SELECT goto FROM alias WHERE address='%s' AND a... tdepasse
52+ hosts = 127.0.0.1
53+ dbname = postfix
54+ <strong class="search_hit">query</strong> = SELECT domain FROM domain WHERE domain='%s'
55+ #optional <strong class="search_hit">query</strong> to use when relaying for backup MX
56+ #<strong class="search_hit">query</strong> = SELECT domain FROM domain WHERE domain='%s' and backupmx =</dd><dt><a href="/tutoriel/comment_creer_un_terminal_x_ou_recycler_une_vieille_machine?s[]=query" class="wikilink1" title="tutoriel:comment_creer_un_terminal_x_ou_recycler_une_vieille_machine">tutoriel:comment_creer_un_terminal_x_ou_recycler_une_vieille_machine</a>: 13 Occurrences trouvées</dt><dd>z gdm (ubuntu) tapez sudo /etc/init.d/gdm stop
57+X -<strong class="search_hit">query</strong> 192.168.1.2
58+&lt;/code&gt;
59+:)
60+Si vous désirez, sur la mê... ans une console (tjs sur le vieil ordi)
61+&lt;code&gt;
62+X -<strong class="search_hit">query</strong> 192.168.1.2 :1
63+&lt;/code&gt;
64+Un écran de login devrait ... ure.
65+&lt;note tip&gt;Rajouter "-once" à la commande "X -<strong class="search_hit">query</strong> 192.168.1.2 :1" permet de quitter la session et r... d'une ubuntu/kubuntu\\
66+Testez d'abord que le //X -<strong class="search_hit">query</strong> ...// fonctionne, dans une console (CTRL-ALT-F1) </dd>
67+ <dt><a href="/bind9?s[]=query" class="wikilink1" title="bind9">bind9</a>: 12 Occurrences trouvées</dt>
68+ <dd> printcmd
69+;; Got answer:
70+;; -&gt;&gt;HEADER&lt;&lt;- opcode: <strong class="search_hit">QUERY</strong>, status: NOERROR, id: 13427
71+;; flags: qr aa rd ra; <strong class="search_hit">QUERY</strong>: 1, ANSWER: 1, AUTHORITY: 1, ADDITIONAL: 1
72+
73+[...]
74+
75+;; <strong class="search_hit">Query</strong> time: 1 msec
76+;; SERVER: 127.0.0.1#53(127.0.0.1)
77+;... ne énorme diminution du temps mis par la requête (<strong class="search_hit">Query</strong> time) , entre la première et la deuxième requête.</dd>
78+ </dl>
79+ """
80+ response = mock.Mock(text=html)
81+ results = doku.response(response)
82+ self.assertEqual(type(results), list)
83+ self.assertEqual(len(results), 4)
84+ self.assertEqual(results[0]['title'], 'xvnc')
85+# FIXME self.assertEqual(results[0]['url'], u'http://this.should.be.the.link/ű')
86+# FIXME self.assertEqual(results[0]['content'], 'This should be the content.')