Sebastian Thiel

I don't know what changed exactly or why the tests would fail now, but judging by the log the failures seem to be due to some foundational issue. I recommend...

Indeed, it's not possible to copy-paste most of it, but from what I can tell it explains itself and the peculiarities coming with being runnable in the test-suite. I do...

Thanks for chiming me in - I was too busy in the past days but am glad I join now since all information has been gathered so neatly. It took...

> it's UB automatically, regardless of what it allows Uninitialised memory is even meaner than I thought 😅. I will prepare a new release now.

I am probably not following on some intricacies of the problem, but wonder if it's possible to create a custom `Write/Read` that is `Clone` and uses shared ownership internally. That...

Thanks for reaching out, even though it's probably close to impossible to tell what's going on without spending considerable time to get going here. As for the error message, I...

Thanks for the fun catchup! I take away that somehow, a flush package remains in the underlying reader, possibly paired with a progress message prior to that. What's strange is...

[Recently I have updated code](https://github.com/Byron/gitoxide/commit/95847091dc7e37ae6cafb07d698c27661953d530) that may be relevant to this issue. *Is this something you could test easily?* Another question came to mind when thinking about this issue, *namely...

Thanks so much for trying it out, it would surely have been nice if it would just have worked :). And since there is no custom transport I'd assume that...

I have posted a fix for this [in the related PR](https://github.com/radicle-dev/heartwood/pull/556) over at `heartblood`, which indeed has to be made on the call-side. In short, it's the same fix but...