Torrents.csv

Demo Server

Torrents.csv is a collaborative repository of torrents and their files, consisting of a searchable torrents.csv, and torrent_files.json. With it you can search for torrents, or files within torrents. It aims to be a universal file system for popular data.

Its initially populated with a January 2017 backup of the pirate bay, and new torrents are periodically added from various torrents sites. It comes with a self-hostable Torrents.csv webserver, a command line search, and a folder scanner to add torrents, and their files.

Torrents.csv will only store torrents with at least one seeder to keep the file small, will be periodically purged of non-seeded torrents, and sorted by infohash.

To request more torrents, or add your own, go here.

Made with Rust, ripgrep, Actix, Inferno, Typescript.

Webserver

Torrents.csv comes with a simple webserver. Demo Server

Requirements

• Rust
• Yarn
• SQLite3
• jq command line JSON parser: Needs at least jq-1.6

Running

git clone https://gitlab.com/dessalines/torrents.csv
cd torrents.csv/scripts
./webserver.sh

and goto http://localhost:8080

If running on a different host, run export TORRENTS_CSV_ENDPOINT=http://whatever.com to change the hostname.

The torrent data is updated daily, and to do so, run, or place this in a crontab:

cd scripts && ./git_update.sh

This updates the repository, and rebuilds the sqlite cache necessary for searching.

Docker

git clone https://gitlab.com/dessalines/torrents.csv
cd torrents.csv
docker-compose up

Command Line Searching

Requirements

• ripgrep

Running

git clone https://gitlab.com/dessalines/torrents.csv
cd torrents.csv
./search.sh "bleh season 1"
bleh season 1 (1993-)
	seeders: 33
	size: 13GiB
	link: magnet:?xt=urn:btih:INFO_HASH_HERE

Uploading / Adding Torrents from a Directory

An upload, consists of making a pull request after running the scan_torrents.sh script, which adds torrents from a directory you choose to the .csv file, after checking that they aren't already there, and that they have seeders. It also adds their files to torrent_files.json.

Requirements

• Torrent-Tracker-Health Dessalines branch
  • npm i -g dessalines/torrent-tracker-health
• jq command line JSON parser: Needs at least jq-1.6
• NodeJS
• Gnu Parallel

Running

Click here to fork this repo.

git clone https://gitlab.com/[MY_USER]/torrents.csv
cd torrents.csv/scripts
./scan_torrents.sh MY_TORRENTS_DIR # `MY_TORRENTS_DIR` is `~/.local/share/data/qBittorrent/BT_backup/` for qBittorrent on linux, but you can search for where your torrents are stored for your client.
git commit -am "Adding [MY_USER] torrents"
git push

Then click here to do a pull/merge request to my branch.

Web scraping torrents

Torrents.csv has a Rust repository for scraping new and top torrents from some torrent sites in the new_torrents_fetcher folder. It currently scrapes skytorrents, magnetdl, and leetx.

Requirements

• Rust
• Cloudflare Scrape
  • sudo pip install cfscrape

Running

git clone https://gitlab.com/dessalines/torrents.csv
cd torrents.csv/scripts
./update.sh SAVE_TORRENT_DIR

API

A JSON output of search results is available at:

http://localhost:8080/service/search?q=[QUERY]&size=[NUMBER_OF_RESULTS]&offset=[PAGE]

How the torrents.csv file looks

infohash;name;size_bytes;created_unix;seeders;leechers;completed;scraped_date
# torrents here...

How the torrent_files.json looks

{
  "012ae083ec82bf911f4fe503b9f6df1effaad9ac": [
    {
      "i": 0, // the index
      "p": "File 1", // the path
      "l": 88546036A // the size in bytes
    },
    ...
  ]
}