git.iko.soy/searx
on master help sign in
Add Tokyo Toshokan search engine
commit547b8a87653d87b8be85710275a66be1bec1e39c authorKirill Isakov <ukwt@ya.ru> dateSun 27 Mar 2016, 00:49 (11 years ago) parent7fbc12e
3 files changed+218
compare with master
unifiedsplit
searx/engines/tokyotoshokan.py+102 searx/settings.yml+6 tests/unit/engines/test_tokyotoshokan.py+110
searx/engines/tokyotoshokan.pynew file+102
@@ -0,0 +1,102 @@
1+"""
2+ Tokyo Toshokan (A BitTorrent Library for Japanese Media)
3+
4+ @website https://www.tokyotosho.info/
5+ @provide-api no
6+ @using-api no
7+ @results HTML
8+ @stable no (HTML can change)
9+ @parse url, title, publishedDate, seed, leech,
10+ filesize, magnetlink, content
11+"""
12+
13+import re
14+from cgi import escape
15+from urllib import urlencode
16+from lxml import html
17+from searx.engines.xpath import extract_text
18+from datetime import datetime
19+from searx.engines.nyaa import int_or_zero, get_filesize_mul
20+
21+# engine dependent config
22+categories = ['files', 'videos', 'music']
23+paging = True
24+
25+# search-url
26+base_url = 'https://www.tokyotosho.info/'
27+search_url = base_url + 'search.php?{query}'
28+
29+
30+# do search-request
31+def request(query, params):
32+ query = urlencode({'page': params['pageno'],
33+ 'terms': query})
34+ params['url'] = search_url.format(query=query)
35+ return params
36+
37+
38+# get response from search-request
39+def response(resp):
40+ results = []
41+
42+ dom = html.fromstring(resp.text)
43+ rows = dom.xpath('//table[@class="listing"]//tr[contains(@class, "category_0")]')
44+
45+ # check if there are no results or page layout was changed so we cannot parse it
46+ # currently there are two rows for each result, so total count must be even
47+ if len(rows) == 0 or len(rows) % 2 != 0:
48+ return []
49+
50+ # regular expression for parsing torrent size strings
51+ size_re = re.compile('Size:\s*([\d.]+)(TB|GB|MB|B)', re.IGNORECASE)
52+
53+ # processing the results, two rows at a time
54+ for i in xrange(0, len(rows), 2):
55+ # parse the first row
56+ name_row = rows[i]
57+
58+ links = name_row.xpath('./td[@class="desc-top"]/a')
59+ params = {
60+ 'template': 'torrent.html',
61+ 'url': links[-1].attrib.get('href'),
62+ 'title': extract_text(links[-1])
63+ }
64+ # I have not yet seen any torrents without magnet links, but
65+ # it's better to be prepared to stumble upon one some day
66+ if len(links) == 2:
67+ magnet = links[0].attrib.get('href')
68+ if magnet.startswith('magnet'):
69+ # okay, we have a valid magnet link, let's add it to the result
70+ params['magnetlink'] = magnet
71+
72+ # no more info in the first row, start parsing the second one
73+ info_row = rows[i + 1]
74+ desc = extract_text(info_row.xpath('./td[@class="desc-bot"]')[0])
75+ for item in desc.split('|'):
76+ item = item.strip()
77+ if item.startswith('Size:'):
78+ try:
79+ # ('1.228', 'GB')
80+ groups = size_re.match(item).groups()
81+ multiplier = get_filesize_mul(groups[1])
82+ params['filesize'] = int(multiplier * float(groups[0]))
83+ except Exception as e:
84+ pass
85+ elif item.startswith('Date:'):
86+ try:
87+ # Date: 2016-02-21 21:44 UTC
88+ date = datetime.strptime(item, 'Date: %Y-%m-%d %H:%M UTC')
89+ params['publishedDate'] = date
90+ except Exception as e:
91+ pass
92+ elif item.startswith('Comment:'):
93+ params['content'] = item
94+ stats = info_row.xpath('./td[@class="stats"]/span')
95+ # has the layout not changed yet?
96+ if len(stats) == 3:
97+ params['seed'] = int_or_zero(extract_text(stats[0]))
98+ params['leech'] = int_or_zero(extract_text(stats[1]))
99+
100+ results.append(params)
101+
102+ return results
searx/settings.yml+6
@@ -271,6 +271,12 @@ engines:
271271 shortcut : sw
272272 disabled : True
273273
274+ - name : tokyotoshokan
275+ engine : tokyotoshokan
276+ shortcut : tt
277+ timeout : 6.0
278+ disabled : True
279+
274280 - name : torrentz
275281 engine : torrentz
276282 timeout : 5.0
tests/unit/engines/test_tokyotoshokan.pynew file+110
@@ -0,0 +1,110 @@
1+import mock
2+from collections import defaultdict
3+from searx.engines import tokyotoshokan
4+from searx.testing import SearxTestCase
5+from datetime import datetime
6+
7+
8+class TestTokyotoshokanEngine(SearxTestCase):
9+
10+ def test_request(self):
11+ query = 'test_query'
12+ dic = defaultdict(dict)
13+ dic['pageno'] = 1
14+ params = tokyotoshokan.request(query, dic)
15+ self.assertTrue('url' in params)
16+ self.assertTrue(query in params['url'])
17+ self.assertTrue('tokyotosho.info' in params['url'])
18+
19+ def test_response(self):
20+ resp = mock.Mock(text='<html></html>')
21+ self.assertEqual(tokyotoshokan.response(resp), [])
22+
23+ html = """
24+ <table class="listing">
25+ <tbody>
26+ <tr class="shade category_0">
27+ <td rowspan="2">
28+ <a href="/?cat=7"><span class="sprite_cat-raw"></span></a>
29+ </td>
30+ <td class="desc-top">
31+ <a href="magnet:?xt=urn:btih:4c19eb46b5113685fbd2288ed2531b0b">
32+ <span class="sprite_magnet"></span>
33+ </a>
34+ <a rel="nofollow" type="application/x-bittorrent" href="http://www.nyaa.se/f">
35+ Koyomimonogatari
36+ </a>
37+ </td>
38+ <td class="web"><a rel="nofollow" href="details.php?id=975700">Details</a></td>
39+ </tr>
40+ <tr class="shade category_0">
41+ <td class="desc-bot">
42+ Authorized: <span class="auth_ok">Yes</span>
43+ Submitter: <a href="?username=Ohys">Ohys</a> |
44+ Size: 10.5MB |
45+ Date: 2016-03-26 16:41 UTC |
46+ Comment: sample comment
47+ </td>
48+ <td style="color: #BBB; font-family: monospace" class="stats" align="right">
49+ S: <span style="color: red">53</span>
50+ L: <span style="color: red">18</span>
51+ C: <span style="color: red">0</span>
52+ ID: 975700
53+ </td>
54+ </tr>
55+
56+ <tr class="category_0">
57+ <td rowspan="2">
58+ <a href="/?cat=7"><span class="sprite_cat-raw"></span></a>
59+ </td>
60+ <td class="desc-top">
61+ <a rel="nofollow" type="application/x-bittorrent" href="http://google.com/q">
62+ Owarimonogatari
63+ </a>
64+ </td>
65+ <td class="web"><a rel="nofollow" href="details.php?id=975700">Details</a></td>
66+ </tr>
67+ <tr class="category_0">
68+ <td class="desc-bot">
69+ Submitter: <a href="?username=Ohys">Ohys</a> |
70+ Size: 932.84EB |
71+ Date: QWERTY-03-26 16:41 UTC
72+ </td>
73+ <td style="color: #BBB; font-family: monospace" class="stats" align="right">
74+ S: <span style="color: red">0</span>
75+ </td>
76+ </tr>
77+ </tbody>
78+ </table>
79+ """
80+
81+ resp = mock.Mock(text=html)
82+ results = tokyotoshokan.response(resp)
83+
84+ self.assertEqual(type(results), list)
85+ self.assertEqual(len(results), 2)
86+
87+ # testing the first result, which has correct format
88+ # and should have all information fields filled
89+ r = results[0]
90+ self.assertEqual(r['url'], 'http://www.nyaa.se/f')
91+ self.assertEqual(r['title'], 'Koyomimonogatari')
92+ self.assertEqual(r['magnetlink'], 'magnet:?xt=urn:btih:4c19eb46b5113685fbd2288ed2531b0b')
93+ self.assertEqual(r['filesize'], int(1024 * 1024 * 10.5))
94+ self.assertEqual(r['publishedDate'], datetime(2016, 03, 26, 16, 41))
95+ self.assertEqual(r['content'], 'Comment: sample comment')
96+ self.assertEqual(r['seed'], 53)
97+ self.assertEqual(r['leech'], 18)
98+
99+ # testing the second result, which does not include magnet link,
100+ # seed & leech info, and has incorrect size & creation date
101+ r = results[1]
102+ self.assertEqual(r['url'], 'http://google.com/q')
103+ self.assertEqual(r['title'], 'Owarimonogatari')
104+
105+ self.assertFalse('magnetlink' in r)
106+ self.assertFalse('filesize' in r)
107+ self.assertFalse('content' in r)
108+ self.assertFalse('publishedDate' in r)
109+ self.assertFalse('seed' in r)
110+ self.assertFalse('leech' in r)