浏览代码

Add Nyaa.se search engine

Kirill Isakov 9 年前
父节点
当前提交
e5677ae6b6
共有 3 个文件被更改,包括 185 次插入0 次删除
  1. 115
    0
      searx/engines/nyaa.py
  2. 4
    0
      searx/settings.yml
  3. 66
    0
      tests/unit/engines/test_nyaa.py

+ 115
- 0
searx/engines/nyaa.py 查看文件

@@ -0,0 +1,115 @@
1
+"""
2
+ Nyaa.se (Anime Bittorrent tracker)
3
+
4
+ @website      http://www.nyaa.se/
5
+ @provide-api  no
6
+ @using-api    no
7
+ @results      HTML
8
+ @stable       no (HTML can change)
9
+ @parse        url, title, content, seed, leech, torrentfile
10
+"""
11
+
12
+from cgi import escape
13
+from urllib import urlencode
14
+from lxml import html
15
+from searx.engines.xpath import extract_text
16
+
17
+# engine dependent config
18
+categories = ['files', 'images', 'videos', 'music']
19
+paging = True
20
+
21
+# search-url
22
+base_url = 'http://www.nyaa.se/'
23
+search_url = base_url + '?page=search&{query}&offset={offset}'
24
+
25
+# xpath queries
26
+xpath_results = '//table[@class="tlist"]//tr[contains(@class, "tlistrow")]'
27
+xpath_category = './/td[@class="tlisticon"]/a'
28
+xpath_title = './/td[@class="tlistname"]/a'
29
+xpath_torrent_file = './/td[@class="tlistdownload"]/a'
30
+xpath_filesize = './/td[@class="tlistsize"]/text()'
31
+xpath_seeds = './/td[@class="tlistsn"]/text()'
32
+xpath_leeches = './/td[@class="tlistln"]/text()'
33
+xpath_downloads = './/td[@class="tlistdn"]/text()'
34
+
35
+
36
+# convert a variable to integer or return 0 if it's not a number
37
+def int_or_zero(num):
38
+    if isinstance(num, list):
39
+        if len(num) < 1:
40
+            return 0
41
+        num = num[0]
42
+    if num.isdigit():
43
+        return int(num)
44
+    return 0
45
+
46
+
47
+# do search-request
48
+def request(query, params):
49
+    query = urlencode({'term': query})
50
+    params['url'] = search_url.format(query=query, offset=params['pageno'])
51
+    return params
52
+
53
+
54
+# get response from search-request
55
+def response(resp):
56
+    results = []
57
+
58
+    dom = html.fromstring(resp.text)
59
+
60
+    for result in dom.xpath(xpath_results):
61
+        # category in which our torrent belongs
62
+        category = result.xpath(xpath_category)[0].attrib.get('title')
63
+
64
+        # torrent title
65
+        page_a = result.xpath(xpath_title)[0]
66
+        title = escape(extract_text(page_a))
67
+
68
+        # link to the page
69
+        href = page_a.attrib.get('href')
70
+
71
+        # link to the torrent file
72
+        torrent_link = result.xpath(xpath_torrent_file)[0].attrib.get('href')
73
+
74
+        # torrent size
75
+        try:
76
+            file_size, suffix = result.xpath(xpath_filesize)[0].split(' ')
77
+
78
+            # convert torrent size to bytes.
79
+            # if there is no correct index in this dictionary,
80
+            # the try block fails as it should
81
+            multiplier = {
82
+                'KIB': 1024,
83
+                'MIB': 1024 ** 2,
84
+                'GIB': 1024 ** 3,
85
+                'TIB': 1024 ** 4
86
+            }[suffix.upper()]
87
+
88
+            file_size = int(float(file_size) * multiplier)
89
+        except Exception as e:
90
+            file_size = None
91
+
92
+        # seed count
93
+        seed = int_or_zero(result.xpath(xpath_seeds))
94
+
95
+        # leech count
96
+        leech = int_or_zero(result.xpath(xpath_leeches))
97
+
98
+        # torrent downloads count
99
+        downloads = int_or_zero(result.xpath(xpath_downloads))
100
+
101
+        # content string contains all information not included into template
102
+        content = 'Category: "{category}". Downloaded {downloads} times.'
103
+        content = content.format(category=category, downloads=downloads)
104
+        content = escape(content)
105
+
106
+        results.append({'url': href,
107
+                        'title': title,
108
+                        'content': content,
109
+                        'seed': seed,
110
+                        'leech': leech,
111
+                        'filesize': file_size,
112
+                        'torrentfile': torrent_link,
113
+                        'template': 'torrent.html'})
114
+
115
+    return results

+ 4
- 0
searx/settings.yml 查看文件

@@ -175,6 +175,10 @@ engines:
175 175
     engine : mixcloud
176 176
     shortcut : mc
177 177
 
178
+  - name : nyaa
179
+    engine : nyaa
180
+    shortcut : nt
181
+
178 182
   - name : openstreetmap
179 183
     engine : openstreetmap
180 184
     shortcut : osm

+ 66
- 0
tests/unit/engines/test_nyaa.py 查看文件

@@ -0,0 +1,66 @@
1
+from collections import defaultdict
2
+import mock
3
+from searx.engines import nyaa
4
+from searx.testing import SearxTestCase
5
+
6
+
7
+class TestNyaaEngine(SearxTestCase):
8
+
9
+    def test_request(self):
10
+        query = 'test_query'
11
+        dic = defaultdict(dict)
12
+        dic['pageno'] = 1
13
+        params = nyaa.request(query, dic)
14
+        self.assertTrue('url' in params)
15
+        self.assertTrue(query in params['url'])
16
+        self.assertTrue('nyaa.se' in params['url'])
17
+
18
+    def test_response(self):
19
+        resp = mock.Mock(text='<html></html>')
20
+        self.assertEqual(nyaa.response(resp), [])
21
+
22
+        html = """
23
+        <table class="tlist">
24
+          <tbody>
25
+            <tr class="trusted tlistrow">
26
+              <td class="tlisticon">
27
+                <a href="//www.nyaa.se" title="English-translated Anime">
28
+                   <img src="//files.nyaa.se" alt="English-translated Anime">
29
+                </a>
30
+              </td>
31
+              <td class="tlistname">
32
+                <a href="//www.nyaa.se/?page3">
33
+                  Sample torrent title
34
+                </a>
35
+              </td>
36
+              <td class="tlistdownload">
37
+                <a href="//www.nyaa.se/?page_dl" title="Download">
38
+                  <img src="//files.nyaa.se/www-dl.png" alt="DL">
39
+                </a>
40
+              </td>
41
+              <td class="tlistsize">10 MiB</td>
42
+              <td class="tlistsn">1</td>
43
+              <td class="tlistln">3</td>
44
+              <td class="tlistdn">666</td>
45
+              <td class="tlistmn">0</td>
46
+            </tr>
47
+          </tbody>
48
+        </table>
49
+        """
50
+
51
+        resp = mock.Mock(text=html)
52
+        results = nyaa.response(resp)
53
+
54
+        self.assertEqual(type(results), list)
55
+        self.assertEqual(len(results), 1)
56
+
57
+        r = results[0]
58
+        self.assertTrue(r['url'].find('www.nyaa.se/?page3') >= 0)
59
+        self.assertTrue(r['torrentfile'].find('www.nyaa.se/?page_dl') >= 0)
60
+        self.assertTrue(r['content'].find('English-translated Anime') >= 0)
61
+        self.assertTrue(r['content'].find('Downloaded 666 times.') >= 0)
62
+
63
+        self.assertEqual(r['title'], 'Sample torrent title')
64
+        self.assertEqual(r['seed'], 1)
65
+        self.assertEqual(r['leech'], 3)
66
+        self.assertEqual(r['filesize'], 10 * 1024 * 1024)