Spark: the API server for Nexirift. Please see the Disclaimer before using.
This is a recode of the original internal Spark server. Even though this is a second revision, it is still not ideal for production use (but has less messy code than the first revision). Only use it once we deem that it is production ready (aka: when Nexirift publicly releases). By running this software in production, you understand the risks.
- KeyDB
- Or a Redis compatible client, we no longer recommend official Redis.
- If you would like to know why we don't suggest it, read the comments here.
- Authentik
- support for more authentication providers is in the works!
psst: see our Authentik stuff here.
Our documentation website is still a work in progress, but you might find better success with guides on there as they should be updated regularly. You can visit that here.
- Set up the prerequistes first.
- Clone the project by using Git:
git clone https://github.com/Nexirift/spark. - Install packages using bun:
bun install. - Start the server using:
bun dev. - Configure the
.envvalues to fit your needs. - Use a tool like Insomnia to send a GQL request!
There is no public docker image (for now), so you have to build it yourself.
- Have docker and docker compose set up.
- Clone the project by using Git:
git clone https://github.com/Nexirift/spark. - Run the stack by doing
docker compose up -d.- This should auto build the image.
You can fill in these values to login to an authentication provider, such as Authentik:
http://authentik.local/application/o/authorize/?
client_id=clientidgoeshere
&redirect_uri= https://openidconnect.net/callback
&scope=openid profile email
&response_type=code
