Sebastian Thiel

One day when the `ein` tool is more useful to more people, I think that definitely would be a nice thing to have. It could probably reuse a lot from...

Thanks a lot for putting this feature on the radar permanently, and for starting the discussion about it. > * I'm not seeing anything like this [verification of blobs/fsck] today...

> (conveniently, this is available in `gix` CLI): That's clever - I forgot `gix tree entries` is even a thing 😅, especially one that can display the sizes of objects....

I think there are a couple of things worth untangling, from my perspective: * connectivity checks aren't strictly needed to implement a first step towards partial clone support * algorithms...

Thanks so much for the update - exciting! I love that you already cooked up a version of the connectivity check and that it apparently runs very fast, too! Too...

> I'm not sure if this is cheating or not... I had expected that git wouldn't be doing any sort of auto-refresh from disk during operations that walk a commit...

Email sent: ``` To whom it may concern, I am the main author of Gitoxide ([https://github.com/Byron/gitoxide](https://github.com/Byron/gitoxide/issues/1028)) and the maintainer of GitPython. Is there anything I need from you to be...

Reply received 2023-09-24: The request will be on the agenda next time the project leadership meets as they make the decisions. A response may be a couple of weeks away...

I'd definitely want to have something like `TreeUpdateBuilder` at some point as well, if only for completeness. I am saying this as I think that an `index.to_tree()` may happen before...

Thanks for reporting! `git` seems to be able to do it by writing packed-refs right away. ``` # pack-refs with: peeled fully-peeled sorted 2f3c410f1900acbf8bb3d29f008d1f9b57103053 refs/remotes/origin/foo