qBittorrent search engine plugins
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

265 lines
9.8 KiB

# VERSION: 1.0
# AUTHORS: imDMG
# NoNaMe-Club search engine plugin for qBittorrent
import tempfile
import os
import logging
import json
import time
from urllib.request import build_opener, HTTPCookieProcessor, ProxyHandler
from urllib.parse import urlencode # , parse_qs
from urllib.error import URLError, HTTPError
from http.cookiejar import Cookie, CookieJar
from html.parser import HTMLParser
from novaprinter import prettyPrinter
# setup logging into qBittorrent/logs
logging.basicConfig(level=logging.DEBUG,
format='%(asctime)s %(name)-12s %(levelname)-8s %(message)s',
datefmt='%m-%d %H:%M',
filename=os.path.abspath(os.path.join(os.path.dirname(__file__), '../../logs', 'nnmclub.log')),
filemode='w')
# benchmark
start_time = time.time()
class nnmclub(object):
name = 'NoNaMe-Club'
url = 'https://nnm-club.me/forum/'
supported_categories = {'all': '-1',
'movies': '14',
'tv': '27',
'music': '16',
'games': '17',
'anime': '24',
'software': '21'}
# getting config from kinozal.json
config = None
try:
# try to load user data from file
with open(os.path.abspath(os.path.join(os.path.dirname(__file__), 'nnmclub.json'))) as f:
config = json.load(f)
except OSError as e:
# file not found
logging.error(e)
raise e
def __init__(self):
# establish connection
#
# make cookie
cj = CookieJar()
# if we wanna use https we mast add ssl=enable_ssl to cookie
c = Cookie(0, 'ssl', "enable_ssl", None, False, '.nnm-club.me',
True, False, '/', True, False, None, 'ParserCookie', None, None, None)
cj.set_cookie(c)
self.session = build_opener(HTTPCookieProcessor(cj))
# add proxy handler if needed
if self.config['proxy'] and any(self.config['proxies'].keys()):
self.session.add_handler(ProxyHandler(self.config['proxies']))
# change user-agent
self.session.addheaders.pop()
self.session.addheaders.append(('User-Agent', self.config['ua']))
response = self._catch_error_request(self.url + 'login.php')
parser = self.WorstParser(self.url, True)
parser.feed(response.read().decode('cp1251'))
parser.close()
form_data = {"username": self.config['username'],
"password": self.config['password'],
"autologin": "on",
"code": parser.login_code,
"login": "Вход"}
# so we first encode keys to cp1251 then do default decode whole string
data_encoded = urlencode({k: v.encode('cp1251') for k, v in form_data.items()}).encode()
self._catch_error_request(self.url + 'login.php', data_encoded)
if 'phpbb2mysql_4_sid' not in [cookie.name for cookie in cj]:
logging.warning("we not authorized, please check your credentials")
else:
logging.info('We successfully authorized')
class WorstParser(HTMLParser):
def __init__(self, url='', login=False):
HTMLParser.__init__(self)
self.url = url
self.login = login
self.torrent = {'link': '',
'name': '',
'size': '',
'seeds': '',
'leech': '',
'desc_link': '', }
self.login_code = None
# we need a page markup to know when stop and collect data,
# because available methods, in this class, do not communicate each other
# as a result, we make markup to transfer information
# from one method to another, along a chain
#
# markup on result table
self.result_table = False # table with results is found
self.torrent_row = False # found torrent row for collect data
self.index_td = 0 # td counter in torrent row
self.write = None # trigger to detecting when to collect data
# markup pagination
self.paginator = False # found more pages in result
self.pages = 0 # page counter
self.search_id = 0
self.found_torrents = 0
def handle_starttag(self, tag, attrs):
# login
if self.login and tag == 'input':
tmp = dict(attrs)
if tmp.get('name') == 'code':
self.login_code = tmp['value']
return
# search result table by class tablesorter
if tag == 'table':
for name, value in attrs:
if name == 'class' and 'tablesorter' in value:
self.result_table = True
# search for torrent row by class prow
if self.result_table and tag == 'tr':
for name, value in attrs:
if name == 'class' and 'prow' in value:
self.torrent_row = True
# count td for find right td
if self.torrent_row and tag == 'td':
if self.index_td == 5:
self.write = "size"
elif self.index_td == 7:
self.write = "seeds"
elif self.index_td == 8:
self.write = "leech"
self.index_td += 1
# search for torrent link by classes r0 or r1
if self.torrent_row and tag == 'a':
if self.index_td == 3:
self.torrent['desc_link'] = self.url + attrs[1][1]
self.write = "name"
if self.index_td == 5:
self.torrent['link'] = self.url + attrs[0][1]
# search for right div with class paginator
if self.found_torrents == 50 and tag == 'span':
for name, value in attrs:
if name == 'class' and value == 'nav':
self.paginator = True
# search for block with page numbers
if self.paginator and tag == 'a':
# if not self.pages:
# parsing for search_id
# self.search_id = parse_qs(attrs[0][1].split('?')[1])['search_id']
self.pages += 1
def handle_endtag(self, tag):
# detecting that torrent row is closed and print all collected data
if self.torrent_row and tag == 'tr':
self.torrent["engine_url"] = self.url
logging.debug('torrent row: ' + str(self.torrent))
prettyPrinter(self.torrent)
self.torrent = {key: '' for key in self.torrent}
self.index_td = 0
self.torrent_row = False
self.found_torrents += 1
# detecting that table with result is close
if self.result_table and tag == 'table':
self.result_table = False
# detecting that we found all pagination
if self.paginator and tag == 'span':
self.paginator = False
def handle_data(self, data: str):
# detecting that we need write data at this moment
if self.write and self.result_table:
if data.startswith('<b>'):
data = data[3:-5]
if self.index_td == 5:
data = data.split('</u>')[1].strip()
self.torrent[self.write] = data.strip()
self.write = None
def error(self, message):
pass
def download_torrent(self, url):
# Create a torrent file
file, path = tempfile.mkstemp('.torrent')
file = os.fdopen(file, "wb")
# Download url
response = self._catch_error_request(url)
# Write it to a file
file.write(response.read())
file.close()
# return file path
logging.debug(path + " " + url)
print(path + " " + url)
def search(self, what, cat='all'):
c = self.supported_categories[cat]
query = '{}tracker.php?nm={}&{}'.format(self.url, what.replace(" ", "+"), "f=-1" if c == '-1' else "c=" + c)
response = self._catch_error_request(query)
parser = self.WorstParser(self.url)
parser.feed(response.read().decode('cp1251'))
parser.close()
# if first request return that we have pages, we do cycle
if parser.pages:
for x in range(1, parser.pages):
response = self._catch_error_request('{}&start={}'.format(query, # &search_id=
# parser.search_id,
parser.found_torrents))
parser.feed(response.read().decode('cp1251'))
parser.close()
6 years ago
logging.info("Found torrents: {}".format(parser.found_torrents))
def _catch_error_request(self, url='', data=None):
url = url if url else self.url
try:
response = self.session.open(url, data)
# Only continue if response status is OK.
if response.getcode() != 200:
logging.error('Unable connect')
raise HTTPError(response.geturl(), response.getcode(),
"HTTP request to {} failed with status: {}".format(url, response.getcode()),
response.info(), None)
except (URLError, HTTPError) as e:
logging.error(e)
raise e
return response
if __name__ == "__main__":
nnmclub_se = nnmclub()
nnmclub_se.search('supernatural')
print("--- %s seconds ---" % (time.time() - start_time))