GoChat server is a learning Go project. It might be completed with a TUI app to also have a client part.
As a learning project, no maintenance is planned yet.
As a learning feature, I don't plan password forgotten features and similar. Just look use cases to see what's in the scope of planned features.
Regarding the architecture I plan to stick to something relatively simple. However I did not have definitive answers yet. At the moment here are the points I think are gonna be true:
- It is a server/client architecture, opposed to P2P (peer-to-peer) ;
- It is going to use http(s) transport for synchronous communication ;
- It is going to use some Socket-like transport for real-time.
To store data I'll start with a relationnal database. Regarding the code organization I think I'll follow some hexagonal architecture principals.
To be able to build the project, you need to have go installed. On Arch you can typically install it like so:
sudo pacman -S goThen just build with the Go command:
go buildTo run the server you can either run it via the Go command:
go run .or via the built executable (have to be built earlier):
./gochatIf you're having trouble with permissions, ensure you have the executable right:
sudo chmod +x ./gochatAlso you can start a PostgreSQL database container with this docker command:
docker run --name gochat-server-db -e POSTGRES_PASSWORD=password -p 5432:5432 -d postgresThis way you can run the server like so:
DATABASE_URL="postgresql://postgres:password@localhost:5432/postgres?sslmode=disable" ./gochat-serverKeep in mind this is for local development of course, not production ready.
Simply run
go testTo build the Docker image just use the Docker command:
docker build .In this section we will see how to install a GoChat server to actually use it in a production way.
To manually install the application you can actually use the Go command:
go installNote that it will install the binary in the GOBIN path. Make sure this is in your system's shell path if you want to use it simply with gochat command. More information in the related Go tutorial.
You can either build the image or fetch one from the GitHub repository. This section still has to be completed later once I'll have at least one Docker image published.