config.py

RSS = [
    { 'url': 'https://example.com' },
    { 'url': 'https://example.com', 'regex': '^.+720p.*$' },
]

PROXIES = {
    'http': 'http://example.com:8080',
    'https': 'https://example.com:8443',
}

TORRENT_DIR = 'torrents'
DRYRUN = False

rsstorrent.py

#!/usr/bin/env python3

import feedparser
import os
import re
import requests
import subprocess
import json
import config
from datetime import datetime


try:
    with open('.timestamps', 'r') as _file:
        timestamps = json.load(_file)
except FileNotFoundError:
    timestamps = dict()


def get_torrent(session, url):
    filename = url.split('/')[-1]
    os.makedirs(config.TORRENT_DIR, exist_ok=True)
    try:
        torrent = session.get(url, timeout=5, proxies=config.PROXIES)
    except (requests.exceptions.Timeout, requests.exceptions.ProxyError):
        print('Request timeout: {}'.format(url))
        return
    with open(f'{config.TORRENT_DIR}/{filename}', 'wb') as _file:
        _file.write(torrent.content)


def get_items(parsed_feed, feed):
    items = list()
    if 'regex' not in feed:
        feed['regex'] = ''
    for item in parsed_feed['entries']:
        if re.match(feed['regex'], item['title']):
            if 'date_parsed' in item:
                items.append({'date': datetime(*item['date_parsed'][:6]), 'title': item['title'], 'link': item['link']})
            else:
                items.append({'date': datetime(*item['published_parsed'][:6]), 'title': item['title'], 'link': item['link']})
    return items


def get_lastitemdate(feed):
    date = int(datetime(2000, 1, 1).timestamp())
    if feed['url'] in timestamps:
        if 'regex' in feed and feed['regex'] in timestamps[feed['url']]:
            date = timestamps[feed['url']][feed['regex']]
        else:
            date = timestamps[feed['url']]['']
    return date


def set_lastitemdate(items, feed):
    last_item_date = int(items[0]['date'].timestamp())
    if feed['url'] not in timestamps:
        timestamps[feed['url']] = dict()
    if 'regex' in feed:
        timestamps[feed['url']][feed['regex']] = last_item_date
    else:
        timestamps[feed['url']][''] = last_item_date
    with open('.timestamps', 'w') as _file:
        json.dump(timestamps, _file)


def download(session, items, feed):
    if len(items) == 0:
        return
    last_item_date = get_lastitemdate(feed)
    for item in items:
        item_date = int(item['date'].timestamp())
        if item_date > last_item_date:
            print('Downloading: {}'.format(item['title']))
            if item['link'].startswith('http'):
                if config.DRYRUN == False:
                    get_torrent(session, item['link'])
            else:
                with open(os.devnull, 'w') as dev_null:
                    if config.DRYRUN == False:
                        subprocess.run(['transmission-remote', '--add', item['link']], stdout=dev_null)
    if config.DRYRUN == False:
        set_lastitemdate(items, feed)


def parse(session, feed):
    try:
        response = session.get(feed['url'], timeout=5, proxies=config.PROXIES)
    except (requests.exceptions.Timeout, requests.exceptions.ProxyError):
        print('Request timeout: {}'.format(feed['url']))
        return
    parsed_feed = feedparser.parse(response.text)
    if parsed_feed['bozo'] == 0:
        items = get_items(parsed_feed, feed)
        download(session, items, feed)
    else:
        print('Bad feed: {}'.format(feed['url']))


if __name__ == '__main__':
    for feed in config.RSS:
        with requests.Session() as session:
            parse(session, feed)