commit: 489bf0f6fd1b153d20aad2ca34706cc214d91558
parent: 700baf2178d5970cef7cd1efaf24522a458a6b55
Author: Adam Tauber <asciimoo@gmail.com>
Date: Wed, 30 Aug 2017 21:08:31 +0200
Merge pull request #1009 from misnyo/torrentz
[fix]torrentz search engine fixed for new version
Diffstat:
3 files changed, 42 insertions(+), 34 deletions(-)
diff --git a/searx/engines/torrentz.py b/searx/engines/torrentz.py
@@ -1,7 +1,7 @@
"""
- Torrentz.eu (BitTorrent meta-search engine)
+ Torrentz2.eu (BitTorrent meta-search engine)
- @website https://torrentz.eu/
+ @website https://torrentz2.eu/
@provide-api no
@using-api no
@@ -14,24 +14,24 @@
import re
from lxml import html
from datetime import datetime
-from searx.engines.nyaa import int_or_zero, get_filesize_mul
from searx.engines.xpath import extract_text
from searx.url_utils import urlencode
+from searx.utils import get_torrent_size
# engine dependent config
categories = ['files', 'videos', 'music']
paging = True
# search-url
-# https://torrentz.eu/search?f=EXAMPLE&p=6
-base_url = 'https://torrentz.eu/'
+# https://torrentz2.eu/search?f=EXAMPLE&p=6
+base_url = 'https://torrentz2.eu/'
search_url = base_url + 'search?{query}'
# do search-request
def request(query, params):
page = params['pageno'] - 1
- query = urlencode({'q': query, 'p': page})
+ query = urlencode({'f': query, 'p': page})
params['url'] = search_url.format(query=query)
return params
@@ -54,22 +54,29 @@ def response(resp):
# extract url and remove a slash in the beginning
link = links[0].attrib.get('href').lstrip('/')
- seed = result.xpath('./dd/span[@class="u"]/text()')[0].replace(',', '')
- leech = result.xpath('./dd/span[@class="d"]/text()')[0].replace(',', '')
+ seed = 0
+ leech = 0
+ try:
+ seed = int(result.xpath('./dd/span[4]/text()')[0].replace(',', ''))
+ leech = int(result.xpath('./dd/span[5]/text()')[0].replace(',', ''))
+ except:
+ pass
params = {
'url': base_url + link,
'title': title,
- 'seed': int_or_zero(seed),
- 'leech': int_or_zero(leech),
+ 'seed': seed,
+ 'leech': leech,
'template': 'torrent.html'
}
# let's try to calculate the torrent size
try:
- size_str = result.xpath('./dd/span[@class="s"]/text()')[0]
- size, suffix = size_str.split()
- params['filesize'] = int(size) * get_filesize_mul(suffix)
+ filesize_info = result.xpath('./dd/span[3]/text()')[0]
+ filesize, filesize_multiplier = filesize_info.split()
+ filesize = get_torrent_size(filesize, filesize_multiplier)
+
+ params['filesize'] = filesize
except:
pass
@@ -80,9 +87,8 @@ def response(resp):
# extract and convert creation date
try:
- date_str = result.xpath('./dd/span[@class="a"]/span')[0].attrib.get('title')
- # Fri, 25 Mar 2016 16:29:01
- date = datetime.strptime(date_str, '%a, %d %b %Y %H:%M:%S')
+ date_ts = result.xpath('./dd/span[2]')[0].attrib.get('title')
+ date = datetime.fromtimestamp(float(date_ts))
params['publishedDate'] = date
except:
pass
diff --git a/searx/settings.yml b/searx/settings.yml
@@ -552,6 +552,12 @@ engines:
timeout : 6.0
disabled : True
+ - name : torrentz
+ engine : torrentz
+ shortcut : tor
+ url: https://torrentz2.eu/
+ timeout : 3.0
+
- name : twitter
engine : twitter
shortcut : tw
diff --git a/tests/unit/engines/test_torrentz.py b/tests/unit/engines/test_torrentz.py
@@ -14,7 +14,7 @@ class TestTorrentzEngine(SearxTestCase):
params = torrentz.request(query, dic)
self.assertTrue('url' in params)
self.assertTrue(query in params['url'])
- self.assertTrue('torrentz.eu' in params['url'])
+ self.assertTrue('torrentz2.eu' in params['url'])
def test_response(self):
resp = mock.Mock(text='<html></html>')
@@ -30,13 +30,11 @@ class TestTorrentzEngine(SearxTestCase):
books ebooks
</dt>
<dd>
- <span class="v">1</span>
- <span class="a">
- <span title="Sun, 22 Nov 2015 03:01:42">4 months</span>
- </span>
- <span class="s">30 MB</span>
- <span class="u">14</span>
- <span class="d">1</span>
+ <span>1</span>
+ <span title="1503595924">5 hours</span>
+ <span>30 MB</span>
+ <span>14</span>
+ <span>1</span>
</dd>
</dl>
@@ -48,13 +46,11 @@ class TestTorrentzEngine(SearxTestCase):
books ebooks
</dt>
<dd>
- <span class="v">1</span>
- <span class="a">
- <span title="Sun, 2124091j0j190gm42">4 months</span>
- </span>
- <span class="s">30MB</span>
- <span class="u">5,555</span>
- <span class="d">1,234,567</span>
+ <span>1</span>
+ <span title="1503595924 aaa">5 hours</span>
+ <span>30MB</span>
+ <span>5,555</span>
+ <span>1,234,567</span>
</dd>
</dl>
</div>
@@ -68,10 +64,10 @@ class TestTorrentzEngine(SearxTestCase):
# testing against the first result
r = results[0]
- self.assertEqual(r['url'], 'https://torrentz.eu/4362e08b1d80e1820fb2550b752f9f3126fe76d6')
+ self.assertEqual(r['url'], 'https://torrentz2.eu/4362e08b1d80e1820fb2550b752f9f3126fe76d6')
self.assertEqual(r['title'], 'Completely valid info books ebooks')
# 22 Nov 2015 03:01:42
- self.assertEqual(r['publishedDate'], datetime(2015, 11, 22, 3, 1, 42))
+ self.assertEqual(r['publishedDate'], datetime.fromtimestamp(1503595924))
self.assertEqual(r['seed'], 14)
self.assertEqual(r['leech'], 1)
self.assertEqual(r['filesize'], 30 * 1024 * 1024)
@@ -79,7 +75,7 @@ class TestTorrentzEngine(SearxTestCase):
# testing against the second result
r = results[1]
- self.assertEqual(r['url'], 'https://torrentz.eu/poaskdpokaspod')
+ self.assertEqual(r['url'], 'https://torrentz2.eu/poaskdpokaspod')
self.assertEqual(r['title'], 'Invalid hash and date and filesize books ebooks')
self.assertEqual(r['seed'], 5555)
self.assertEqual(r['leech'], 1234567)
