Sebastian Thiel

In the end it comes down to a `Transport` implementation which, combined [with a `Remote`](https://docs.rs/gix/latest/gix/struct.Remote.html#method.to_connection_with_transport) becomes a connection. The linked method already takes a trait object. This kind of setting...

TBH, doing this would be ideal and it would be great if that would be possible. The downstream would have to decide that though as the async versions of the...

@djc It definitely is also me who doesn't like `async` and the `Pin` handling whenever one has to implement a trait. It's just not nice compared to Sync code, so...

That would basically be the proposed solution 1: ditch async, and async impls could be used under the hood. It's all possible, but I think this must be driven by...

That sounds very reasonable and I'd love to get a chance to clean up this age-old tech-debt. Internally there already is a clear separation, so it's more of a refactoring...

Just for completeness, here is [another comment](https://github.com/GitoxideLabs/gitoxide/issues/2214#issue-3506937618), which is reproduced here for convenience: > In the gitoxide Cargo workspace, there are a number of features that violate the requirement that...

Thanks a lot. If using `open` in its place is backwards compatible within the v3 version of python, this should be an easy fix.

Thanks for reporting. I haven't used this myself, so I wouldn't know. But if it's not working it might be a bug in `clap` which is the one generating the...

Thanks for reporting and the initial investigation, and sorry for the hassle. I understand this issue isn't present before #2155 was merged, and that it's unrelated to the pack index...

Thanks so much for chiming in @folkertdev! As a note, I think disabling `max-performance` shouldn't make a difference (anymore) as by now we always use `zlib-rs`, and we always use...