Next episodes downloader from http://www.nyaa.eu/

nya.py

#!/usr/bin/env python2
# -*- coding: utf-8 -*-
import re
import sys
import os.path

from urllib2 import build_opener
from urllib import urlencode
from lxml import html

DLDIR = '/home/bobrov/downloads'
TRDIR = u'/home/bobrov/torrents'
SEARCH_URL = 'http://www.nyaa.eu/?{}'
TEAMS = ('HorribleSubs', 'Commie', 'Zero-Raws', 'Leopard-Raws')

SD_TEAMS = ('Commie', )

opener = build_opener()

def dump(resp, fname='/tmp/boo.html'):
    with open(fname, 'w') as f:
        f.write(resp.read())

def get_team_idx(title, teams):
    for i, t in enumerate(teams):
        if t in title:
            return i

    return -1

def get_torrent_url_for_next_episode(fname):
    title, _, episode = fname.lstrip('! ').rpartition(' - ')
    if title.startswith('['):
        teams, _, title = title.partition(']')
        title = title.strip()
        teams = map(str.strip, teams.lstrip('[').split('|'))
    else:
        teams = TEAMS

    episode = '{:0{}}'.format(int(episode) + 1, len(episode))

    query = '{} {}'.format(title, episode)
    params = {
        'page': 'search',
        'cats': '0_0',
        'filter': '2',
        'term': query
    }

    resp = opener.open(SEARCH_URL.format(urlencode(params)))
    root = html.parse(resp)

    candidates = []
    for tr in root.xpath('//tr[contains(@class, "tlistrow")]'):
        title = tr.xpath('./td[@class="tlistname"]/a')[0].text.replace('_', ' ')
        if ('720' in title or any(r in title for r in SD_TEAMS)) and ('- '+episode) in title:
            tidx = get_team_idx(title, teams)
            if tidx >= 0:
                seeders = int(tr.xpath('./td[@class="tlistsn"]')[0].text)
                if seeders < 10:
                    continue

                turl = tr.xpath('./td[@class="tlistdownload"]/a')[0].attrib['href']
                candidates.append((tidx, title, turl))

    if candidates:
        return min(candidates)[2]
    else:
        return None

def fetch_torrent(url):
    resp = opener.open(url)
    fname = re.search('filename="(.+?)"', resp.headers['Content-Disposition']).group(1).decode('utf-8')

    if any(os.path.exists(os.path.join(TRDIR, fname + r)) for r in (u'', u'.added')):
        print fname, 'already exists'
    else:
        print fname, 'added'
        dump(resp, os.path.join(TRDIR, fname))

def main():
    if len(sys.argv) > 1:
        fnames = sys.argv[1:]
    else:
        fnames = [fname for fname in os.listdir(DLDIR) if fname.startswith('!!!')]

    for fname in fnames:
        url = get_torrent_url_for_next_episode(fname)
        if url:
            fetch_torrent(url)

if __name__ == '__main__':
    main()