triatk/nyaarr repository overview

⁠Nyaarr

Nyaarr is a bridge between Nyaa Anime and Sonarr, playing an important role to convert the anime release titles to normalized sonarr format.

This document only describes how to use the project.

⁠Introduction

Sonarr⁠ based on RSS feeds to find releases (Release) that meet your needs and download them, It identifies the corresponding episodes, languages, and formats based on the title (Title) of the RSS feed. However, the identification algorithm and episode mapping is not perfect and cant be customized, especially for Japanese animation translation, which is poorly standardized. Therefore, we rewrite the titles to ensure that it can be correctly recognized by Sonarr.

This process requires the user to provide the template format of the title and the corresponding language and media format information. In order to facilitate management, we provide a simple front-end/back-end system.

⁠Use

First you need a functioning and accessible Sonarr instance, and get the API Key⁠.

Next, create the .env file:

SONARR_API_KEY=aaaabbbbccccddddeeeeffff1145141919810
SONARR_HOST=https://sonarr.yourdomain.com
ADMIN_USERNAME=nyaarr
ADMIN_PASSWORD=your_admin_password
BASE_URL=https://nyaa.si

ADMIN_USERNAME and ADMIN_PASSWORD are used to log in to the system, Nyaarr cannot be accessed without logging in.

Then create a jwk key under the data directory:

ssh-keygen -t rsa -b 4096 -E SHA512 -m PEM -f jwt.key
openssl rsa -in jwt.key -pubout -outform PEM -out jwt.key.pub

Then run this project (requires Node.js environment):

$ yarn build
$ yarn start

Will start a server on port 12307.

At this point, you can visit http://localhost:12307 to enter the Nyaarr web page:

In the web page, you can search, sort, add, delete, and edit items. The main fields that need to be filled include:

• Remote: Remote address, that is, the RSS feed address of Nyaa Anime, optional.
• Pattern: title template, only items matching at least one item will be returned, the template itself is a regular expression⁠, the expression needs to contain name capture Groups⁠ (i.e. (?<episode>\d+) ) for episode information.
• Series: series name, the web page will get all series names from Sonarr, if the desired series is not found, you may need to add the series in Sonarr first.
• Season: Season code, the web side will get all seasons from Sonarr and display their monitoring status.
• Language: language, which is the English name of the language corresponding to the episode, such as Chinese, which needs to meet the language settings required by the series, otherwise it cannot be captured by Sonarr.
• Quality: quality, available values ​​can refer to Sonarr source code⁠, Sonarr doesn't seem to support it The quality of the items pushed by RSS is automatically detected, and the default value on the web page is WEBDL 1080p.

In create and edit pages, the sidebar will automatically fetch the items in the RSS feed based on the address in the Remote field, and highlight the items that match the template:

Note that the content of Pattern is regular expression, and the common [] in the seed title is a special character in the regular expression, which needs to be escaped, you can click the ESCAPE button after the input box Quickly escape the contents of the input box, or click the EPISODE button to quickly copy the episode's regex rule.

For example, if the episode title you want to grab is:

[Lilith-Raws] If Ultimate Evolved Fully Immersive RPGs Are More Like Junk Games Than Reality / Full Dive - 03 [Baha][WEB-DL][1080p][AVC AAC][CHT][MP4]

Paste it into the input box and hit ESCAPE, you will get:

\[Lilith-Raws\] If Ultimate Evolution's fully immersive RPG is more of a trash game than reality/ Full Dive - 03 \[Baha\]\[WEB-DL\]\[1080p\]\[AVC AAC\ ]\[CHT\]\[MP4\]

Then click EPISODE , replace the 03 in the box with the copied content, and get:

\[Lilith-Raws\] If Ultimate Evolution's fully immersive RPG is more of a garbage game than reality/ Full Dive - (?<episode>\d+) \[Baha\]\[WEB-DL\]\[1080p \]\[AVC AAC\]\[CHT\]\[MP4\]

Fill in the other fields correctly, and you have completed editing a grab template.

Of course, a more convenient method is to find the corresponding RSS feed on Nyaa Anime, fill in the Remote field, and select the desired entry in the sidebar to complete the first two steps.

To get the transformed RSS feed, you can replace the domain name part of Nyaa Anime (nyaa.si) directly with the domain name deployed by Nyaarr, for example from:

https://nyaa.si/?page=rss&c=1_3&f=0

get:

https://<Nyaarr domain name>/RSS/?page=rss&c=1_3&f=0

Add this address to Sonarr and it will automatically grab the episodes you want, cool?

If you fill in the Remote field, you can also click the PROXY button to copy it with one click.

⁠Other

⁠Deployment

You can deploy using Docker⁠, our Docker Image is at triatk/nyaarr⁠ .

The resulting image does not contain a .env file, you need to put it in the data/ folder below.

Also, data files should not be included in Docker Image and Docker Container, you should mount the data folder data/ from the host to /data in the container.

If you use docker-compose , here is a docker-compose.yml template:

version: "3"

services:
  nyaarr:
    image: triatk/nyaarr
    volumes:
      - /path/on/host/data:/data
    environment:
      - PUID=1000
      - PGID=100
    expose:
      - 12307/tcp
    restart: unless-stopped

This will start the application using the same .env file.

⁠Safety

• Paths starting with /RSS will be used to convert RSS feeds.
• /proxy is used to allow the front end to request Nyaa Anime's RSS feed, only URLs under the https://nyaa.si domain name are allowed to be requested.
• /sonarr is used to reverse the request to the Sonarr API to avoid the client saving the API Key.
• Paths starting with /api are used to access and manipulate data.
• Other addresses will get the corresponding static files under build/, if not found, it will get index.html . Since the frontend uses the Hash Router, all frontend HTML requests should only access the / path.

Considering the simplicity of implementation, we do not have access control on the server, you can add access control based on the above description, such as [HTTP Basic Auth](https://developer.mozilla.org/en-US/docs/⁠ Web/HTTP/Authentication). The path under /?page=rss needs to be accessed by Sonarr, HTTP Auth cannot be used, and user data will not be directly exposed, so no access control can be considered.

⁠Development

We use the same package.json to cover both front-end and back-end code, which is not good practice, but convenient.

Beyond that, the following files make up a basic create-react-app⁠ application:

src/
public/
build/ (build gets)

The following files are only used by the backend code:

server/
data/ (data file)
.env (configuration file)

For front-end development, you can use yarn dev:web , which uses react-scripts dev to start a Dev Server locally.

In order to access the backend API, front-end development requires a running backend server, you can use yarn dev:server or directly yarn start to start one. We assume that the backend address for development is http://localhost:12307 , you can also change the proxy field in package.json to change.

yarn dev:server is also used to develop backend code. It will use nodemon⁠ to listen to files in server/ and automatically restart the server when the file changes. A running instance of the frontend is not required to develop the backend.

After the build is completed, the back-end server will serve the front-end code directly, and then only yarn start is needed to start one copy.

⁠Author

Xavier Xiong - Homepage⁠

张海川 - Haichuan Zhang - [email protected]⁠ - Homepage⁠