eightball/PluralKit
1
0
Fork 0
mirror of https://github.com/PluralKit/PluralKit.git synced 2026-02-04 04:56:49 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

chore: update readme, example config, and running documentation

Browse source 
Co-authored-by: asleepyskye <skye@vixen.lgbt>
This commit is contained in:
alyssa 2025-04-24 10:42:40 +00:00
parent 9506833abf
commit 437ea72ed4
6 changed files with 96 additions and 67 deletions
Download patch file Download diff file Expand all files Collapse all files

69
README.md
View file

@ -5,75 +5,14 @@ PluralKit is a Discord bot meant for plural communities. It has features like me
PluralKit has a Discord server for support, feedback, and discussion: https://discord.gg/PczBt78 PluralKit has a Discord server for support, feedback, and discussion: https://discord.gg/PczBt78
# Requirements
Running the bot requires [.NET 5](https://dotnet.microsoft.com/download), a PostgreSQL database and a Redis database. It should function on any system where the prerequisites are set up (including Windows).
Optionally, it can integrate with [Sentry](https://sentry.io/welcome/) for error reporting and [InfluxDB](https://www.influxdata.com/products/influxdb-overview/) for aggregate statistics.
# Configuration
Configuring the bot is done through a JSON configuration file. An example of the configuration format can be seen in [`pluralkit.conf.example`](https://github.com/PluralKit/PluralKit/blob/master/pluralkit.conf.example).
The configuration file needs to be placed in the bot's working directory (usually the repository root) and must be called `pluralkit.conf`.
The configuration file is in JSON format (albeit with a `.conf` extension). The following keys are available (using `.` to indicate a nested object level), bolded key names are required:
* **`PluralKit.Bot.Token`**: the Discord bot token to connect with
* **`PluralKit.Database`**: the URI of the database to connect to (in [ADO.NET Npgsql format](https://www.connectionstrings.com/npgsql/))
* **`PluralKit.RedisAddr`**: the `host:port` of a Redis database to connect to
* `PluralKit.Bot.Prefixes`: an array of command prefixes to use (default `["pk;", "pk!"]`).
* **`PluralKit.Bot.ClientId`**: the ID of the bot's user account, used for calculating the bot's own permissions and for the link in `pk;invite`.
* `PluralKit.SentryUrl` *(optional)*: the [Sentry](https://sentry.io/welcome/) client key/DSN to report runtime errors to. If absent, disables Sentry integration.
* `PluralKit.InfluxUrl` *(optional)*: the URL to an [InfluxDB](https://www.influxdata.com/products/influxdb-overview/) server to report aggregate statistics to. An example of these stats can be seen on [the public stats page](https://stats.pluralkit.me).
* `PluralKit.InfluxDb` *(optional)*: the name of an InfluxDB database to report statistics to. If either this field or `PluralKit.InfluxUrl` are absent, InfluxDB reporting will be disabled.
* `PluralKit.LogDir` *(optional)*: the directory to save information and error logs to. If left blank, will default to `logs/` in the current working directory.
The bot can also take configuration from environment variables, which will override the values read from the file. Here, use `:` (colon) or `__` (double underscore) as a level separator (eg. `export PluralKit__Bot__Token=foobar123`) as per [ASP.NET config](https://docs.microsoft.com/en-us/aspnet/core/fundamentals/configuration/?view=aspnetcore-3.1#environment-variables).
# Running # Running
See [RUNNING.md](./RUNNING.md).
## Docker # Development
The easiest way to get the bot running is with Docker. The repository contains a `docker-compose.yml` file ready to use. See [the dev-docs/ directory](./dev-docs/README.md)
* Clone this repository: `git clone https://github.com/PluralKit/PluralKit`
* Create a `pluralkit.conf` file in the same directory as `docker-compose.yml` containing at least `PluralKit.Bot.Token` and `PluralKit.Bot.ClientId` fields
* (`PluralKit.Database` is overridden in `docker-compose.yml` to point to the Postgres container)
* Build the bot: `docker-compose build`
* Run the bot: `docker-compose up`
In other words:
```
$ git clone https://github.com/PluralKit/PluralKit
$ cd PluralKit
$ cp pluralkit.conf.example pluralkit.conf
$ nano pluralkit.conf # (or vim, or whatever)
$ docker-compose up -d
```
## Manually
* Install the .NET 6 SDK (see https://dotnet.microsoft.com/download)
* Clone this repository: `git clone https://github.com/PluralKit/PluralKit`
* Create and fill in a `pluralkit.conf` file in the same directory as `docker-compose.yml`
* Run the bot: `dotnet run --project PluralKit.Bot`
* Alternatively, `dotnet build -c Release -o build/`, then `dotnet build/PluralKit.Bot.dll`
(tip: use `scripts/run-test-db.sh` to run a temporary PostgreSQL database on your local system. Requires Docker.)
## Scheduled Tasks worker
There is a scheduled tasks worker that needs to be ran separately from the bot. This handles cleaning up the database, and updating statistics (system/member/etc counts, shown in the `pk;stats` embed).
Note: This worker is *not required*, and the bot will function correctly without it.
If you are running the bot via docker-compose, this is set up automatically.
If you run the bot manually you can run the worker as such:
* `dotnet run --project PluralKit.ScheduledTasks`
* or if you used `dotnet build` rather than `dotnet run` to run the bot: `dotnet build/PluralKit.ScheduledTasks.dll`
# Upgrading database from legacy version
If you have an instance of the Python version of the bot (from the `legacy` branch), you may need to take extra database migration steps.
For more information, see [LEGACYMIGRATE.md](./LEGACYMIGRATE.md).
# User documentation # User documentation
See [the docs/ directory](./docs/README.md) See [the docs/ directory](./docs/README.md)
# License # License
This project is under the GNU Affero General Public License, Version 3. It is available at the following link: https://www.gnu.org/licenses/agpl-3.0.en.html This project is under the GNU Affero General Public License, Version 3. It is available at the following link: https://www.gnu.org/licenses/agpl-3.0.en.html

0
BRANCHRENAME.md → dev-docs/BRANCHRENAME.md
View file

0
LEGACYMIGRATE.md → dev-docs/LEGACYMIGRATE.md
View file

21
dev-docs/README.md Normal file
View file

@ -0,0 +1,21 @@
# PluralKit development documentation
Most of PluralKit's code is written in C#, and is split into 3 projects: `PluralKit.Core` (supporting libraries), `PluralKit.Bot` (the Discord bot), and `PluralKit.API` (ASP.NET webserver with controllers for most API endpoints).
There is an ongoing effort to port this code to Rust, and we have a handful of crates already. Currently, the main Rust services are `gateway` (connects to Discord to receive events), `api` (handles authentication and rate-limiting for the public API, as well as a couple of private endpoints), and `scheduled_tasks` (background cron job runner, for statistics and miscellaneous cleanup).
At the very least, `PluralKit.Bot` and `gateway` are required for the bot to run. While code still exists to connect to the Discord gateway directly from the C# bot, this is no longer a supported configuration and may break in the future.
Service-specific documentation can be found for the C# services in [dotnet.md](./dotnet.md), and for the Rust services in [rust.md](./rust.md) (todo; there is a temporary mostly-accurate document at [RUNNING.md](./RUNNING.md)).
## Building/running
PluralKit uses a PostgreSQL database and a Redis database to store data. User-provided data is stored in Postgres; Redis is used for internal state and transient data such as the command execution cache. It's generally easy to run these in Docker or with the Nix `process-compose`, but any install method should work.
The production instance of PluralKit uses Docker images built in CI. These take a long time to rebuild and aren't good for development (they're production builds, so it's not possible to hook up a debugger). Instead, it's preferable to install build dependencies locally. This is easy with the provided Nix flake: run `nix develop .#bot` to drop into a shell with all the C# build dependencies available, and `nix develop .#services` to get a shell with the Rust build dependencies. It's also okay to manually install build dependencies if you prefer.
PluralKit services are configured with environment variables; see service-specific documentation for details. Generally, the configuration from the self-host `docker-compose.yml` should get you started.
## Upgrading database from legacy version
If you have an instance of the Python version of the bot (from the `legacy` branch), you may need to take extra database migration steps.
For more information, see [LEGACYMIGRATE.md](./LEGACYMIGRATE.md).

69
dev-docs/RUNNING.md Normal file
View file

@ -0,0 +1,69 @@
# Technical Overview:
PluralKit is composed of several different parts, some of which optional and not needed for the bot to function in a testing environment.
##### Required:
- Bot (*PluralKit.Bot*)
- Gateway
- PostgreSQL Database
- Redis Database
##### Optional:
- API (*PluralKit.API*)
- Scheduled Tasks (*scheduled_tasks*)
*Optionally, it can also integrate with Sentry for error reporting, and InfluxDB for aggregate statistics. In production, we use [VictoriaMetrics](https://victoriametrics.com/) InfluxDB competible endpoint, to query metrics in a Prometheus-compatible format.
The bot and API are built using .NET 8, with other optional components used for scaling (ex. scheduled_tasks) built using Rust.
# Building + Running
**The below procedures are intended for development and testing use only!
Newer versions of the bot have moved to using [Nix](https://nixos.org/) for simplified builds. A docker compose file is also provided, but not recommended as it is not actively maintained.
See [Configuration](./CONFIGURATION.md) for full details on configuration as only the basics will be covered here.
Configuration is done through a JSON configuration file `pluralkit.conf` placed in the bot's working directory. An example of the configuration format can be seen in [pluralkit.conf.example](pluralkit.conf.example).
The minimum configuration needed for the bot to function must include the following:
- **`PluralKit.Bot.Token`**: the Discord bot token to connect with
- **`PluralKit.Bot.ClientId`**: the ID of the bot's user account, used for calculating the bot's own permissions and for the link in `pk;invite`
- **`PluralKit.Database`**: the URI of the PostgreSQL database to connect to (in [ADO.NET Npgsql format](https://www.connectionstrings.com/npgsql/))
- **`PluralKit.RedisAddr`**: the `host:port` of the Redis database to connect to
**When running using Docker, you do not need to specify the Postgres or Redis URLs as these will be overwritten by environment variables in the compose file.**
**When using Nix, the Database URI Username, Password, and Database fields must match what the database was setup with in the `flake.nix` file!**
The bot can also take configuration from environment variables, which will override the values read from the file. Here, use `:` (colon) or `__` (double underscore) as a level separator (eg. `export PluralKit__Bot__Token=foobar123`) as per [ASP.NET config](https://docs.microsoft.com/en-us/aspnet/core/fundamentals/configuration/?view=aspnetcore-3.1#environment-variables).
## Nix (recommended)
The bot, databases, and services are available to run all in one as a Nix flake (using process-compose-flake).
As of the writing of this document, there are a few caveats with the current flake file.
- The database URI in the config must match the username, password, and database specified during database creation (currently `postgres`, `postgres`, and `pluralkit` respectively).
- Not all services are added to the flake file yet.
**Additionally, as of the writing of this document, the `pluralkit-gateway` service reads environment variables to get the bot token and database URLs, so you also need to create a `.env` file in the `PluralKit` directory with the following variables:**
```
pluralkit__db__db_password="postgres"
pluralkit__db__data_db_uri="postgresql://postgres@localhost:5432/pluralkit"
pluralkit__db__data_redis_addr="redis://localhost:6379"
pluralkit__discord__bot_token="BOT_TOKEN_GOES_HERE"
pluralkit__discord__client_id="BOT_CLIENT_ID_GOES_HERE"
pluralkit__discord__client_secret=1
pluralkit__discord__max_concurrency=1
```
**(This should match the username/password/database specified in the flake file and the configuration file)**
*(assuming you already have Git installed, if not, you can start a shell with git by running `nix-shell -p git`)*
1. Clone the repository: `git clone https://github.com/PluralKit/PluralKit`
2. Create a `pluralkit.conf` configuration file in the `PluralKit` directory
- Again, the DB URI parameters must match what's in the `flake.nix` file
3. Create a `.env` configuration file in the `PluralKit` directory *(see above)*
4. Build and run: `nix run .#dev`
- This will download the dependencies, build, and run PluralKit
- If Nix is not setup to allow flakes, you may need to add `--extra-experimental-features nix-command --extra-experimental-features flakes` to the command
- If the `pluralkit-bot` process fails to run, you can restart it by selecting it and pressing `Ctrl-R`
```
[nix-shell:~]$ git clone https://github.com/PluralKit/PluralKit
[nix-shell:~]$ cd PluralKit
[nix-shell:~/PluralKit]$ cp pluralkit.conf.example pluralkit.conf
[nix-shell:~/PluralKit]$ nano pluralkit.conf
[nix-shell:~/PluralKit]$ nano .env
[nix-shell:~/PluralKit]$ nix run .#dev
```

4
pluralkit.conf.example
View file

@ -4,7 +4,7 @@
"Token": "BOT_TOKEN_GOES_HERE", "Token": "BOT_TOKEN_GOES_HERE",
"ClientId": 466707357099884544, "ClientId": 466707357099884544,
}, },
"Database": "Host=localhost;Port=5432;Username=myusername;Password=mypassword;Database=mydatabasename", "Database": "Host=localhost;Port=5432;Username=postgres;Password=postgres;Database=pluralkit",
"RedisAddr": "host:port" "RedisAddr": "127.0.0.1:6379"
} }
} }