Rewrite ws executable in something "not bash"?

[paulotruta]

The WS executable is a simple bash script with a few arguments. For the proof-of-concept part it works well, but as it complicates we might just re-write the entire logic in something more developer-friendly, like GO, Rust, or other interesting tech.

Suggestions and contributions are welcome!

[paulotruta]

As the sysctl repo develops, we'll end up porting the whole of ws executable into it.

[paulotruta]

Want to discuss about two of the possibilities we have for improving ws. Namely, porting its bootstrapping script into something more manageable and less finicky.

Since we are using Go for edgeboxctl, this seems a solid approach for a rewrite of ws.
Now the question: Should we keep this bootstrapping functionality within ws, or port it into edgeboxctl?

Implications of having a "rewrite ws and keep it in the same repo" strategy are that we will keep concerns separated (good), but we'll have to manage things separately (separated build process and whatnot).

Implications of having a "port to edgeboxctl" stategy are that we'll simplify things by having the majority of Edgeboxe's critical functionality code under one same roof, with the disadvantage of having to create support within edgeboxctl source code for running the build process also outside Edgeboxctl main loop.

Personally, I think having a Go project inside this very repo would not be such a bad thing. We can then compile the binary and run it the same way we already do with the bash script, saving us time avoiding changing anything in edgeboxctl because of this (we keep the modularity factor although these two projects are kind of dependant on each other).

Happy to hear your thoughts ;)