Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/apricote/Listory

Track your Spotify listens
https://github.com/apricote/Listory

analytics nestjs nodejs quantified-self react self-hosted spotify typescript

Last synced: 3 months ago
JSON representation

Track your Spotify listens

Awesome Lists containing this project

README 

        


  


      Listory

    

  



  



  
Login with Spotify and Listory will save all tracks you listen to.





  Latest Release

  GitHub branch checks state

  Codecov

  GitHub




  

      

      

      The listens report shows how many songs you have listened to during the last month (and all months prior to that).

    

  

  

    

      

      Want to know which song you heard on your drive to work last tuesday? Now you can!

    

    

      

      Find out what genres (or artists, albums, songs) you listen to the most!

    

  



## Installation



### Creating the Spotify App



To connect to the Spotify API, you need to create a new App on the [Spotify

Developer Dashboard](https://developer.spotify.com/dashboard/applications).



Once the App is created, open the App Overview and click on the "Edit Settings"

button. A new modal should open. In this modal, you need to add a Redirect URI,

so that the login with Spotify actually works. This URL depends on where you

want to host your Listory installation, see also `APP_URL` under

[Configuration / Application](#application). The Redirect URI should be

`$APP_URL/api/v1/auth/spotify/callback`. For the local example URL which is used

in development, this would be `http://localhost:3000/api/v1/auth/spotify/callback`.

If you have your own domain where you want to host Listory, this is probably

something like `https://listory.your-name.com/api/v1/auth/spotify/callback`.



You can add multiple Redirect URIs and all will work.



Keep the tab open, as you will need the _Client ID_ and _Client Secret_ in the

next step.



### Deployment



Listory currently supports two deployment mechanisms: _docker compose_ and

_Kubernetes Helm Chart_.



#### docker compose



There are two `docker compose` files in the repository, for a production

deployment, you want to use [`docker-compose.prod.yml`](./docker-compose.prod.yml).



You can copy this file to your server or whereever you want to run Listory. You

will also need to copy the `.env.sample` file next to the

`docker-compose.prod.yml` file and rename it to `.env`.



Open the `.env` file in an editor and put in the Spotify App _Client ID_ and

_Client Secret_.



Now you can configure Listory how you like by changing the `environment` of the

`listory` service in the docker compose file, or by adding new values in the

`.env` file. For a list of all available options, see section

[Configuration](#configuration).



If you deploy Listory on the public internet, I recommend you to add a

reverse proxy like [Traefik][traefik], which you can configure to

[automatically add TLS certificates][traefik-tls] (putting the S into HTTPS).



[traefik]: https://doc.traefik.io/traefik/getting-started/quick-start/

[traefik-tls]: https://doc.traefik.io/traefik/user-guides/docker-compose/acme-tls/



Once you have set everything up, you can run this command to start Listory:



```

docker compose up --daemon --file docker-compose.prod.yml

```



This will start Listory in the background. Checkout the [docker compose documentation][docker-compose],

to learn how you can work with the containers, for example to restart them or to

read the logs.



[docker-compose]: https://docs.docker.com/compose/reference/



### Helm Chart



We publish a Kubernetes Helm Chart that installs Listory into a Kubernetes cluster.



I have not yet setup publishing to an actual Chart Registry, so if you would like

to use the chart, create an issue and I will set this up properly.



You can find the source code for the Helm Chart under `[charts/listory](./charts/listory/)`.



### Configuration



All configuration must be set as environment variables. Default values are added in **bold**, values that are required are marked with _Required_.



#### Application



- `PORT`: **3000**: Port the webserver will listen on.

- `APP_URL`: **http://localhost:3000**: Public URL of the Application, is used to generate Links.



#### Authentication



- `JWT_SECRET`: _Required_, used to sign the JWTs.

- `JWT_ALGORITHM`: **HS256**: Algorithm used to sign the JWTs. One of `HS256`, `HS384`, `HS512`

- `JWT_EXPIRATION_TIME`: **1d**: Lifetime of signed JWTs. Accepts strings like `1d`, `2h`, `15m`.



#### Spotify



- `SPOTIFY_CLIENT_ID`: _Required_, Spotify App Client ID

- `SPOTIFY_CLIENT_SECRET`: _Required_, Spotify App Client Secret

- `SPOTIFY_FETCH_INTERVAL_SEC`: **60**: Interval for fetching recently listened tracks from Spotify.

- `SPOTIFY_UPDATE_INTERVAL_SEC`: **60**: Interval for updating previously imported music library entities (artist, album, track). Raise this number if you often hit the Spotify API Ratelimit.

- `SPOTIFY_WEB_API_URL`: **https://api.spotify.com/**: Spotify WEB API Endpoint.

- `SPOTIFY_AUTH_API_URL`: **https://accounts.spotify.com/**: Spotify Authentication API Endpoint.

- `SPOTIFY_USER_FILTER`: **""**: If set, only allow Spotify users with these ids to access the app. If empty, allow all users to access the app. Seperate ids with `,` eg.: `231421323123,other_id`.



#### Database



- `DB_HOST`: _Required_, Database host

- `DB_USERNAME`: _Required_, Database username

- `DB_PASSWORD`: _Required_, Database password

- `DB_DATABASE`: _Required_, Database database

- `DB_POOL_MAX`: **50**, max concurrent database connections



#### Sentry



You can use Sentry to automatically detect and report any exceptions thrown.



- `SENTRY_ENABLED`: **false**, Set to `true` to enable Sentry.

- `SENTRY_DSN`: _Required_, but only if `SENTRY_ENABLED` is `true`. The [DSN](https://docs.sentry.io/product/sentry-basics/dsn-explainer/) for your Sentry project.



#### OpenTelemetry



We use OpenTelemetry to provide observability into Listory API.



The metrics will be exposed on a seperate port at `:9464/metrics`. Make sure that this endpoint is not publicly available in your deployment.



Traces will be sent to the specified endpoint.



To use observability tools locally, check out `docker-compose` setup in `observability/`.



- `OTEL_METRICS_ENABLED`: **false**, Set to `true` to activate metrics.

- `OTEL_TRACES_ENABLED`: **false**, Set to `true` to activate traces.

- `OTEL_EXPORTER_OTLP_ENDPOINT`: _Required_, but only if `OTEL_TRACES_ENABLED` is `true`. The endpoint that traces are sent to, see [OpenTelemetry docs](https://github.com/open-telemetry/opentelemetry-js/tree/main/experimental/packages/exporter-trace-otlp-http#configuration-options-as-environment-variables)

- `OTEL_EXPORTER_PROMETHEUS_PORT`: **9464**, Set to configure non-standard port for Prometheus metrics



## Development



### Configure Spotify API Access



Copy the file `.env.sample` to `.env` and add your Spotify API Key.



### Starting the application



We use `docker compose` to provide a full local development environment.



```bash

$ docker compose up

```



You can now access the frontend at `http://localhost:3000` and the API at `http://localhost:3000/api`.



Frontend and API will automatically reload on any code changes.



### REPL Console



You can start the REPL console by starting the normal environment, and then running:



```bash

$ docker compose run console

```



### Observability



If you want to start the observability suite (Metrics & Tracing) locally, you can use the `observability` docker compose profile:



```bash

$ docker compose --profile observability up

```



Grafana is then available at `http://localhost:2345` and all sources are preconfigured.



## Test



```bash

# unit tests

$ npm run test



# e2e tests

$ npm run test:e2e



# test coverage

$ npm run test:cov

```



## License



Listory is [MIT licensed](LICENSE).