Merge pull request #1022 from misnyo/nyaa

nyaa.si fixed

3 files changed, 51 insertions, 52 deletions

diff --git a/searx/engines/nyaa.py b/searx/engines/nyaa.py
index 272c712c4..6a8e598c4 100644
--- a/searx/engines/nyaa.py
+++ b/searx/engines/nyaa.py
@@ -1,7 +1,7 @@
 """
- Nyaa.se (Anime Bittorrent tracker)
+ Nyaa.si (Anime Bittorrent tracker)
 
- @website      http://www.nyaa.se/
+ @website      http://www.nyaa.si/
  @provide-api  no
  @using-api    no
  @results      HTML
@@ -12,50 +12,25 @@
 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 = 'http://www.nyaa.se/'
+base_url = 'http://www.nyaa.si/'
 search_url = base_url + '?page=search&{query}&offset={offset}'
 
 # xpath queries
-xpath_results = '//table[@class="tlist"]//tr[contains(@class, "tlistrow")]'
-xpath_category = './/td[@class="tlisticon"]/a'
-xpath_title = './/td[@class="tlistname"]/a'
-xpath_torrent_file = './/td[@class="tlistdownload"]/a'
-xpath_filesize = './/td[@class="tlistsize"]/text()'
-xpath_seeds = './/td[@class="tlistsn"]/text()'
-xpath_leeches = './/td[@class="tlistln"]/text()'
-xpath_downloads = './/td[@class="tlistdn"]/text()'
-
-
-# convert a variable to integer or return 0 if it's not a number
-def int_or_zero(num):
-    if isinstance(num, list):
-        if len(num) < 1:
-            return 0
-        num = num[0]
-    if num.isdigit():
-        return int(num)
-    return 0
-
-
-# get multiplier to convert torrent size to bytes
-def get_filesize_mul(suffix):
-    return {
-        'KB': 1024,
-        'MB': 1024 ** 2,
-        'GB': 1024 ** 3,
-        'TB': 1024 ** 4,
-
-        'KIB': 1024,
-        'MIB': 1024 ** 2,
-        'GIB': 1024 ** 3,
-        'TIB': 1024 ** 4
-    }[str(suffix).upper()]
+xpath_results = '//table[contains(@class, "torrent-list")]//tr[not(th)]'
+xpath_category = './/td[1]/a[1]'
+xpath_title = './/td[2]/a[last()]'
+xpath_torrent_links = './/td[3]/a'
+xpath_filesize = './/td[4]/text()'
+xpath_seeds = './/td[6]/text()'
+xpath_leeches = './/td[7]/text()'
+xpath_downloads = './/td[8]/text()'
 
 
 # do search-request
@@ -72,25 +47,32 @@ def response(resp):
     dom = html.fromstring(resp.text)
 
     for result in dom.xpath(xpath_results):
+        # defaults
+        filesize = 0
+        magnet_link = ""
+        torrent_link = ""
+
         # category in which our torrent belongs
-        category = result.xpath(xpath_category)[0].attrib.get('title')
+        try:
+            category = result.xpath(xpath_category)[0].attrib.get('title')
+        except:
+            pass
 
         # torrent title
         page_a = result.xpath(xpath_title)[0]
         title = extract_text(page_a)
 
         # link to the page
-        href = page_a.attrib.get('href')
+        href = base_url + page_a.attrib.get('href')
 
-        # link to the torrent file
-        torrent_link = result.xpath(xpath_torrent_file)[0].attrib.get('href')
-
-        # torrent size
-        try:
-            file_size, suffix = result.xpath(xpath_filesize)[0].split(' ')
-            file_size = int(float(file_size) * get_filesize_mul(suffix))
-        except:
-            file_size = None
+        for link in result.xpath(xpath_torrent_links):
+            url = link.attrib.get('href')
+            if 'magnet' in url:
+                # link to the magnet
+                magnet_link = url
+            else:
+                # link to the torrent file
+                torrent_link = url
 
         # seed count
         seed = int_or_zero(result.xpath(xpath_seeds))
@@ -101,6 +83,14 @@ def response(resp):
         # torrent downloads count
         downloads = int_or_zero(result.xpath(xpath_downloads))
 
+        # let's try to calculate the torrent size
+        try:
+            filesize_info = result.xpath(xpath_filesize)[0]
+            filesize, filesize_multiplier = filesize_info.split()
+            filesize = get_torrent_size(filesize, filesize_multiplier)
+        except:
+            pass
+
         # content string contains all information not included into template
         content = 'Category: "{category}". Downloaded {downloads} times.'
         content = content.format(category=category, downloads=downloads)
@@ -110,8 +100,9 @@ def response(resp):
                         'content': content,
                         'seed': seed,
                         'leech': leech,
-                        'filesize': file_size,
+                        'filesize': filesize,
                         'torrentfile': torrent_link,
+                        'magnetlink': magnet_link,
                         'template': 'torrent.html'})
 
     return results
diff --git a/searx/engines/tokyotoshokan.py b/searx/engines/tokyotoshokan.py
index 9a6b5e57d..773212043 100644
--- a/searx/engines/tokyotoshokan.py
+++ b/searx/engines/tokyotoshokan.py
@@ -14,8 +14,8 @@ import re
 from lxml import html
 from searx.engines.xpath import extract_text
 from datetime import datetime
-from searx.engines.nyaa import int_or_zero, get_filesize_mul
 from searx.url_utils import urlencode
+from searx.utils import get_torrent_size, int_or_zero
 
 # engine dependent config
 categories = ['files', 'videos', 'music']
@@ -76,8 +76,7 @@ def response(resp):
                 try:
                     # ('1.228', 'GB')
                     groups = size_re.match(item).groups()
-                    multiplier = get_filesize_mul(groups[1])
-                    params['filesize'] = int(multiplier * float(groups[0]))
+                    params['filesize'] = get_torrent_size(groups[0], groups[1])
                 except:
                     pass
             elif item.startswith('Date:'):
diff --git a/searx/utils.py b/searx/utils.py
index 3df571160..35b20ad83 100644
--- a/searx/utils.py
+++ b/searx/utils.py
@@ -290,6 +290,15 @@ def convert_str_to_int(number_str):
         return 0
 
 
+# convert a variable to integer or return 0 if it's not a number
+def int_or_zero(num):
+    if isinstance(num, list):
+        if len(num) < 1:
+            return 0
+        num = num[0]
+    return convert_str_to_int(num)
+
+
 def is_valid_lang(lang):
     is_abbr = (len(lang) == 2)
     if is_abbr: