impl: SSE

[funwarioisii]

close: #22

I'm planning to checkout from this branch and proceed with the SSE implementation
Additionally:

• Fix the tests
• Complete the implementation related to SSE

[kfischer-okarin]

I realize that with my ClientConnection refactor I made the wrong assumption about the control flow, i.e. which object is in control of application flow

sequenceDiagram
    participant M as main
    participant S as Server
    participant C as ClientConnection

   M->>S: serve
    loop While server is running
        S->>C: read_next_message
        C-->>S: message
        S->>S: process_input
        S->>C: send_message
        C-->>S: OK
    end

Loading

But it's probably more fitting to have something like this

sequenceDiagram
    participant M as main
    participant S as Server
    participant H as MessageHandler

   M->>S: serve
    loop While server is running
        S->>H: handle_message
        H-->>S: response
    end

Loading

where MessageHandler would basically what the server is right now except for the serve method - which is transport agnostic... And then all the Transport logic would be handled by the Server object - and there would be StdIoServer and an HttpSseServer etc

Maybe it would be better to first refactor it to this easier to handle form before starting SSE 🤔

[kfischer-okarin]

Basically like this
kfischer-okarin@610e6cb

Server#process_input became public method MessageHandler#handle_message

just did the initial extract commit to show practically what I mean - didn't touch the tests - but then renaming/moving most of them to equivalent MessageHandler tests should be trivial - they should become quite a bit simpler too - since we wouldn't have to do the Fiber hacks anymore to circumvent the serve loop inside the tests.....

[martinemde]

Hey, just noting that a new spec dropped today and it changed the implementation. I haven't looked closely.