A small Python script badly written made to automate the addition of airing anime on my server, using Tsundere-Raws'RSS page on Nyaa. Originally made to work on a linux server, with JellyFin, Plex or other Streaming service.

• Python 3.7 at least (support is not assured for older versions)
• a qBittorrent Web UI, hosted on the same machine as the script (use qBittorrent-nox (updated) on servers)
• HandBrakeCLI installed on your linux distro (dpkg, not the flatpack version)
• mkvtoolnix, to extract subtitles
• feedparser, a python module
• python-qBittorrent, another python module ​

apt-get install python3
apt-get install qbittorrent-nox
apt-get install handbrake-cli
apt-get install mkvtoolnix
pip3 install feedparser
pip3 install python-qbittorrent
apt-get update && apt-get upgrade

It will check for new episodes in the RSS every 7 minutes and if it finds one, it will download and encode the episode to the selected location.

(exactly 7 minutes to avoid getting tempban from DNS servers or Nyaa, you can choose between encoding or copying the file) ​

This script was intended to work with Tsundere-Raws torrents, giving the script another RSS feed will certainly break the script.

Default encoding settings : (~500MB ouput file size, for a ~24mn 1080p episode)

• Bitrate : 2500kbps x264 Variable Framerate optimized for HTTP streaming
• Audio Bitrate : 512kbps (HandBrake will automatically lower that value to match the bitrate of the actual episode torrent audio track)
• Subtitle : Hardub

you can of course change the encoding settings, please refer to HandBrakeCLI Documentation ​

• Install all dependencies.
• Start qBitrorrent and enable the WebUI if you aren't using qBittorrent-nox.
• Set the torrent download directory to wherever you want
• Download CanalBot.py into a direcroty.
• Create a file named anime_list.txt next to it, then inside the file include animes infos one line per anime, respecting this pattern : rss_keyword, anime_folder_name, anime_name, see the anime_list.txt example (anime_folder_name may be different from anime_name and vice versa).
  • ⚠️ For the keyword, be sure to use a keyword that will work for both the torrent name on nyaa and the video file name itself. Also make sure that the keyword only works for the anime you're adding (otherwise it could add unwanted animes)
  • Any space in the anime_name will be replaced by a dot.
  • Be sure not to use invalid characters for the file name ("/", "\O", "\n", "\r", "\t", "\b", "\v", "\a" and "\f" in linux).
• Now edit CanalBot.py and modify the first variables included in the settings class to match your settings, see Settings below.
• Now run the script in a screen (linux package : screen -S canalbot) on your linux server to avoid linux killing the process, and that's it.
• If the script crashes, just restart it, it won't re-add episodes, or re-encode episode.

My script will respect this file structure (with default settings plus extract_subtitles enabled):

target_folder/
└─ anime_name/
   └─ s1/
      ├─ anime_name.s1e01.suffix.mp4
      ├─ anime_name.s1e02.suffix.mp4
      ├─ anime_name.s1e03.suffix.mp4
      ├─ ...
      └─ subtitles/                       Only if extract_subtitles is set to True
         ├─ anime_name.s1e01.suffix.ass
         ├─ anime_name.s1e02.suffix.ass
         ├─ anime_name.s1e03.suffix.ass
         └─ ...

• delete_torrents_afterwards if set to True will delete permanently torrent file (the downloaded one, not the file encoded or copied) from disk and qBittorrent Web UI.
• auto_encode do what its name does : if set to True, it will automatically encode the file, however if set to False, it will just copy the file to the desired location.
• lang is used to choose the language of subtitles that will be burned in the video (however Tsundere-Raws only provides french subs).
• extract_subtitles if set to True the script will extract subtitles see the file structure.
• suffix define the end of the file. Episode name will be in this format, again, see the file structure.
• quality obviously the video quality of the episodes you want the script to download (1080p OR 720p).
• handbrake_settings are the settings that HandBrakeCLI will use when encoding the file.
• target_directory tells the script where does the episodes will be storen, see the file structure one more time.
• linuxuser specify which linux user should own the files.
• user is the user of the qBittorrent Web UI.
• password is the password of the qBittorrent Web UI.
• webui_link set the link to the qBittorrent Web UI. ​

• You can customise the encoding command if you want, see HandBrakeCLI Documentation.
• You can stop the script at any moment py pressing CTRL + C, otherwise it will run infinitely.