Running pointercrate copies
- PostgreSQL (database used by pointercrate)
- pandoc (to build the markdown documentation and the guidelines)
- diesel-cli (install via
cargo install diesel_cli. Install with only the postgres feature enabled and make sure you have the postgres library installed to link against)
Pointercrate is mainly configured via optional environment variables. It will attempt to load them from an
.env file. These are:
DATABASE_URL: The URL to the postgres database to connect to, including authentication
PORT: The port to run on (defaults to
DOCUMENTATION: The directory with the compiled documentation html files (defaults to
GUIDELINES: The directory with the compiled guidelines html files (defaults to
LIST_SIZE: Size of the main list (defaults to 50)
EXTENDED_LIST_SIZE: Size of the main list + extended list (defaults to 100)
Additionally, you'll need a
.secret file containing the secret to sign access tokens with.
Getting it running
Even though pointercrate no longer uses diesel as it's database driver, it still uses diesel's migration system. To get a database instance running, run
diesel migration run. You might have to mess around with the initial migrations a bit to get them working because they are partially based upon the existing scheme of a very old python version of pointercrate.
Since pointercrate uses
sqlx, compilation requires you to be running a postgres database with the pointercrate schema. This is because
sqlx validates all SQL queries at compile time (syntactically and semantically) by sending them over to a locally running postgres server.
Please remove the adsense and google analytics scripts when hosting your own copy!
While I'm generally OK (in fact, its pretty awesome) with people running their own copies of this code on their own servers (note: running the code. I'm not OK with people copying the content from pointercrate), doing so is completely unsupported from my side beyond these instructions. If you have enough knowledge to be capable to run a server, I fully believe in you to be able to figure out how to get it running from these. Depending on what mood I'm in on any given day, I might offer support over in my discord server.
Furthermore, if you do run a pointercrate copy, I do ask you to remove all references to pointercrate from it and state somewhere that you are not associated with pointercrate in any way or form. I'd appreciate some credit, but it's not required.
I am under no obligation to keep any potential modifications you have made to the code base from working if you decide to merge in upstream changes.
The following people have helped with development of pointercrate, either through code contributions or other things:
- cos8o: Reverse engineered parts of the Geometry Dash source that allows pointercrate to display accurate object counts and level lengths
- zmx and mgostIH and everyone else over in my discord server
- Nimbus: Development of various discord bots integrating with the pointercrate API
- Aquatias and Deltablu: My trusty admins that click checkboxes for me (love you guys)
- and of course the developers of all the dependencies pointercrate uses