Merge pull request #1283 from rinpatch/acgsou-engine

[Feature] Acgsou as a searchable engine
This commit is contained in:
Adam Tauber 2018-04-28 22:42:35 +02:00 committed by GitHub
commit 1a1f9852f1
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 158 additions and 0 deletions

75
searx/engines/acgsou.py Normal file
View File

@ -0,0 +1,75 @@
"""
Acgsou (Japanese Animation/Music/Comics Bittorrent tracker)
@website https://www.acgsou.com/
@provide-api no
@using-api no
@results HTML
@stable no (HTML can change)
@parse url, title, content, seed, leech, torrentfile
"""
from lxml import html
from searx.engines.xpath import extract_text
from searx.url_utils import urlencode
from searx.utils import get_torrent_size, int_or_zero
# engine dependent config
categories = ['files', 'images', 'videos', 'music']
paging = True
# search-url
base_url = 'https://www.acgsou.com/'
search_url = base_url + 'search.php?{query}&page={offset}'
# xpath queries
xpath_results = '//table[contains(@class, "list_style table_fixed")]//tr[not(th)]'
xpath_category = './/td[2]/a[1]'
xpath_title = './/td[3]/a[last()]'
xpath_torrent_links = './/td[3]/a'
xpath_filesize = './/td[4]/text()'
def request(query, params):
query = urlencode({'keyword': query})
params['url'] = search_url.format(query=query, offset=params['pageno'])
return params
def response(resp):
results = []
dom = html.fromstring(resp.text)
for result in dom.xpath(xpath_results):
# defaults
filesize = 0
magnet_link = "magnet:?xt=urn:btih:{}&tr=http://tracker.acgsou.com:2710/announce"
torrent_link = ""
try:
category = extract_text(result.xpath(xpath_category)[0])
except:
pass
page_a = result.xpath(xpath_title)[0]
title = extract_text(page_a)
href = base_url + page_a.attrib.get('href')
magnet_link = magnet_link.format(page_a.attrib.get('href')[5:-5])
try:
filesize_info = result.xpath(xpath_filesize)[0]
filesize = filesize_info[:-2]
filesize_multiplier = filesize_info[-2:]
filesize = get_torrent_size(filesize, filesize_multiplier)
except:
pass
# I didn't add download/seed/leech count since as I figured out they are generated randomly everytime
content = 'Category: "{category}".'
content = content.format(category=category)
results.append({'url': href,
'title': title,
'content': content,
'filesize': filesize,
'magnetlink': magnet_link,
'template': 'torrent.html'})
return results

View File

@ -434,6 +434,12 @@ engines:
shortcut : nt shortcut : nt
disabled : True disabled : True
- name : acgsou
engine : acgsou
shortcut : acg
disabled : True
timeout: 5.0
- name : openairedatasets - name : openairedatasets
engine : json_engine engine : json_engine
paging : True paging : True

View File

@ -0,0 +1,77 @@
from collections import defaultdict
import mock
from searx.engines import acgsou
from searx.testing import SearxTestCase
class TestAcgsouEngine(SearxTestCase):
def test_request(self):
query = 'test_query'
dic = defaultdict(dict)
dic['pageno'] = 1
params = acgsou.request(query, dic)
self.assertTrue('url' in params)
self.assertTrue(query in params['url'])
self.assertTrue('acgsou.com' in params['url'])
def test_response(self):
resp = mock.Mock(text='<html></html>')
self.assertEqual(acgsou.response(resp), [])
html = """
<html>
<table id="listTable" class="list_style table_fixed">
<thead class="tcat">
<tr>
<th axis="string" class="l1 tableHeaderOver">test</th>
<th axis="string" class="l2 tableHeaderOver">test</th>
<th axis="string" class="l3 tableHeaderOver">test</th>
<th axis="size" class="l4 tableHeaderOver">test</th>
<th axis="number" class="l5 tableHeaderOver">test</th>
<th axis="number" class="l6 tableHeaderOver">test</th>
<th axis="number" class="l7 tableHeaderOver">test</th>
<th axis="string" class="l8 tableHeaderOver">test</th>
</tr>
</thead>
<tbody class="tbody" id="data_list">
<tr class="alt1 ">
<td nowrap="nowrap">date</td>
<td><a href="category.html">testcategory</a></td>
<td style="text-align:left;">
<a href="show-torrentid.html" target="_blank">torrentname</a>
</td>
<td>1MB</td>
<td nowrap="nowrap">
<span class="bts_1">
29
</span>
</td>
<td nowrap="nowrap">
<span class="btl_1">
211
</span>
</td>
<td nowrap="nowrap">
<span class="btc_">
168
</span>
</td>
<td><a href="random.html">user</a></td>
</tr>
</tbody>
</table>
</html>
"""
resp = mock.Mock(text=html)
results = acgsou.response(resp)
self.assertEqual(type(results), list)
self.assertEqual(len(results), 1)
r = results[0]
self.assertEqual(r['url'], 'https://www.acgsou.com/show-torrentid.html')
self.assertEqual(r['content'], 'Category: "testcategory".')
self.assertEqual(r['title'], 'torrentname')
self.assertEqual(r['filesize'], 1048576)