mdeluise / ytsms Goto Github PK

[Still under "initial" development, some features may be unstable or change in the future, although database schemas should be stable. A first release version is planned to be packed soon].

YTSMS is a self-hosted youtube subscription management service.
Useful to keep track of new uploaded video of the followed channels, even without having a Google account.

Why? • Features highlights • Quickstart • Configuration • FAQ • Contribute

I've always enjoyed spending some time viewing YouTube videos. The problem is that YouTube keeps track of a lot of data about you.

Ytsms takes care of this problem allowing you to get updated about new videos from followed channels without having a Google account or having it but using the official API (that is still better then using directly the subscription system of the Google account, since you can view the video logged out).

• Add channels to followed list
• Get notified about every new video published by the followed channels
• Filter videos by channel
• Chose between YouTube backend or an Invidious instance
• Dark/Light mode
• 🔜 Create playlists of videos

Installing ytsms is pretty straight forward, in order to do so follow these steps:

1. Create a folder where you want to place all ytsms related files.
2. Inside that folder, create the following files:
   • Create a directory mkdir certificates
   • Give the permission of the directory chmod 777 ./certificates
   • docker-compose.yml:

   version: "3"
   name: ytsms
   services:
     backend:
       image: msdeluise/ytsms-backend:latest
       env_file: backend.env
       depends_on:
         - db
       restart: unless-stopped
       volumes:
         - "certs:/certificates"
       ports:
         - "8080:8080"
   
     db:
       image: mysql:8.0
       restart: always
       env_file: backend.env
       volumes:
         - "./db:/var/lib/mysql"
   
     frontend:
       image: msdeluise/ytsms-frontend:latest
       env_file: frontend.env
       links:
         - backend
       ports:
         - "3000:3000"
       volumes:
         - "certs:/certificates"
   volumes:
     certs:
       driver: local
       driver_opts:
         type: none
         o: bind
         device: ./certificates

   • backend.env:

   #
   # DB
   #
   MYSQL_HOST=db
   MYSQL_PORT=3306
   MYSQL_USERNAME=root
   MYSQL_PSW=root
   MYSQL_ROOT_PASSWORD=root
   MYSQL_DATABASE=bootdb
   
   #
   # JWT
   #
   JWT_SECRET=putTheSecretHere
   JWT_EXP=1
   
   #
   # Server config
   #
   USERS_LIMIT=-1 # <= 0 means no limit
   YOUTUBE_KEY=
   VIDEO_REFRESH=0 0 * * * *
   
   #
   # SSL
   #
   SSL_ENABLED=false
   CERTIFICATE_PATH=/certificates/

   • frontend.env:

   PORT=3000
   API_URL=http://localhost:8080/api
   WAIT_TIMEOUT=10000
   BROWSER=none
   SSL_ENABLED=false
   CERTIFICATE_PATH=/certificates/

3. Run the docker compose file (docker compose -f docker-compose.yml up -d), then the service will be available at localhost:3000, while the REST API will be available at localhost:8080/api (localhost:8080/api/swagger-ui/index.html for the documentation of them).

Run the docker compose file (docker compose -f <file> up -d), then the service will be available at localhost:8080, while the REST API will be available at localhost:8080/api (localhost:8080/api/swagger-ui/index.html for the documentation of them).

The service offers two video retrieval options:

• scraping: videos are retrieved without the need of a YouTube API key, although there are some restrictions
• official YouTube api: video are retrieved using a YouTube API key, which means more metadata can be fetched but less anonymization is provided

In order to choose between one of the retrieval option, the YOUTUBE_KEY property in the configuration file must be filled or left empty.

• How can I get the channel id of a youtube channel?

  You can do it in two ways:

  • go to the channel homepage and run in the javascript console ytInitialData.metadata.channelMetadataRenderer.externalId
  • go to the channel homepage, view the page source and search for the value browse_id

• What distinguishes official youtube api and scraping modes?

  The official youtube api mode offers the following advantages:

  • retrieve the video duration
  • retrieve the channels image thumbnails
  • retrieve more old video

  The scraping mode offers the following advantages:

  • no need to use any YouTube API key
  • no quota restrictions
  • increased anonymization

• How can I create a YouTube API key?

  You can create a key following the official guide.

• Why sometime the Channel Name method for adding a subscription does not works?

  This is a pretty known problem (1, 2, 3, 4). It's caused by the inconsistence of some YouTube channel's custom URL and username parameters.

Feel free to contribute and help improve the repo.

You can submit any of this in the issues section of the repository. Chose the right template and then fill the required info.

If you fix a bug, please follow the [contribution-guideline](https://github.com/MDeLuise/ytsms# contribution-guideline) in order to merge the fix in the repository.

Let's discuss first possible solutions for the development before start working on that, please open a feature request issue.

To fix a bug or create a feature, follow these steps:

1. Fork the repo
2. Create a new branch (git checkout -b awesome-feature)
3. Make changes or add new changes.
4. Commit your changes (git add -A; git commit -m 'Awesome new feature')
5. Push to the branch (git push origin awesome-feature)
6. Create a Pull Request

• Commits should follow the semantic commit specification, although not mandatory.