derivator / tafkars Goto Github PK

I've noticed the project doesn't follow the usual style of a Rust project. I'd like to propose a change in the structure, which will require a bit of rework. The main changes are a root src/ file for a few files like main, lib, etc. Then the crates can live within the root directory as well - though, I've seen examples where this is the case and isn't. As such, I have 2 proposals I'd like to add - flat and structured. I think we'll be okay with flat, but I wanted to add the structured type as an option.

current

├── Cargo.lock
├── Cargo.toml
├── README.md
├── tafkars
│   ├── Cargo.toml
│   ├── README.md
│   └── src
│       ├── comment.rs
│       ├── lib.rs
│       ├── listing.rs
│       └── submission.rs
└── tafkars-lemmy
    ├── Cargo.lock
    ├── Cargo.toml
    ├── README.md
    └── src
        ├── api_translation.rs
        ├── endpoints.rs
        ├── main.rs
        └── web_config.rs

Proposed1 - flat

Pros: Simple, straightforward. Seen in "smaller" projects such as Atuin
Cons: For bigger projects at least, this is a bit uncommon - see Lemmy or Cargo.

├── Cargo.lock
├── Cargo.toml
├── README.md
├── src
│   ├── main.rs
│   ├── lib.rs
├── tafkars
│   ├── Cargo.toml
│   ├── README.md
│   └── src
│       ├── comment.rs
│       ├── lib.rs
│       ├── listing.rs
│       └── submission.rs
└── tafkars-lemmy
    ├── Cargo.lock
    ├── Cargo.toml
    ├── README.md
    └── src
        ├── lib.rs
        ├── api_translation.rs
        ├── endpoints.rs
        └── web_config.rs

Proposed2 - structured

Pros: Allows for more organization if/when project gets bigger (docker folders, integration tests, scripts, etc.). See see Lemmy or Cargo for large-code examples.
Cons: Can add unnecessary complexity, and isn't always necessary. For example, another large-ish project is Lapce which uses a more flat structure like above.

├── Cargo.lock
├── Cargo.toml
├── README.md
├── crates
│   ├── tafkars
│   │   ├── Cargo.toml
│   │   ├── README.md
│   │   └── src
│   │       ├── comment.rs
│   │       ├── lib.rs
│   │       ├── listing.rs
│   │       └── submission.rs
│   └── tafkars-lemmy
│       ├── Cargo.lock
│       ├── Cargo.toml
│       ├── README.md
│       └── src
│           ├── api_translation.rs
│           ├── endpoints.rs
│           ├── main.rs
│           └── web_config.rs
└── src
    ├── lib.rs
    └── main.rs

I see you have no LICENSE file for this project. The default is copyright.

I would suggest releasing the code under the GPL-3.0-or-later or AGPL-3.0-or-later license so that others are encouraged to contribute changes back to your project.

The last element can also be of kind more if there's additional comments beyond the limit.

Hello!
I'm fairly experienced with Rust and I love the idea of (hopefully) using my reddit client on lemmy, so I'd love to help.

I see you have written that people should contact you on how to get started, but I think creating some issues that people can pick up would go a long way.

With that said, please let me know how and where I can help!

If developers of Lemmy apps wanted to flip the current functionality of tafkar from Reddit(primary)+Lemmy(secondary) to Lemmy+Reddit, does that warrant a whole other library, or is there a natural path for this in tafkar?

See for example https://github.com/gkasdorf/memmy/

A problem here is that reddit uses after/before, while Lemmy uses page to specify the page.

Dump some actual reddit API responses for as many different endpoints as possible and test that they can deserialized