panicbit / flif-rs

Apache License 2.0
0 stars 0 forks source link

Proposal: collaboration between Rust projects #11

Open ZoeyR opened 7 years ago

ZoeyR commented 7 years ago

Hi, I recently started working on my own rust-based flif decoder. I started the work a bit before I found this project and had continued to work on it mostly as a learning experience. As of now I think my project has either reached feature parity with this one or is close (feel free to correct me if I'm massively wrong about that though!).

Before I go any further with my project I thought it would be a good idea to discuss the idea of combining our efforts into one project instead of working on things separately. That way we could avoid massive duplication of work (MANIAC decoding looks to be a lot of work to implement). Alternatively, if you are no longer interested in maintaining this project or for other reasons with the projects to remain separate that is fine as well.

If we were to combine our efforts I think these topics would be relevant to discuss:

  1. Which repository shall this live in?
    • this one?
    • mine?
    • create an overarching organization?
  2. What is the design direction for the project?
    • it would be good to have consensus on this if the projects were to join
    • direct port of C++ code or re-implement segments in idiomatic Rust?
    • should encoding be put on the roadmap?

If you wish to keep the repositories separate I believe the only issue that would appear moving forward would be who would end up with control of the flif crate name in the end. Since you got the offer from the current owner of the crate first I think you should have first dibs.

panicbit commented 6 years ago

Sorry to answer to you this late.

I don't really have the time to continue this project. Since there was no spec when I started it, I was going for a direct C++ translation.

I noticed thst you already started on your own implementation by now, which already looks cleaner than mine, so I don't think there's duplicated effort :).

I think the first person to turn up with a mostly working pure Rust implementation should get the crate.

Good luck with your project!