Edward Thomson

I think that makes sense. What would you like to happen if we encounter a symlink that points outside the tree? Stop and return that? I think that feels like...

Sorry for the delay - were you able to make any progress here? I'm surprised to see this failing on an `https` URL. I would expect that there's a problem...

> I currently do not have a MSVC dev environment available so I cannot check this. Would it help you if we started offering prebuilt binaries for Windows? 🤔

I'm a little curious about the approach: instead of thinking about `GIT_ENOTFOUND` as a "supertype" of not found errors and then introducing `GIT_EMISSING` and `GIT_EPROMISED`, should we keep `GIT_ENOTFOUND` to...

> So if a client looks an object up by ID, and it's not found, it's always going to be GIT_ENOTFOUND since we don't know anything more than that about...

What version of cmake? It [claims to honor the deployment target version](https://gitlab.kitware.com/cmake/cmake/-/issues/17101) for `futimens`

Thanks for the report! > I'd like to check with you whether this is something that should be fixed (i.e. whether you want to support worktrees with slashes), Is there...

> As far as I can see, the name of the directory in .git/worktrees just does not matter. I agree. I don't know if this was a git implementation detail...

> That also means that the change is backwards compatible, as NULL is currently not a valid value for name. Also, any name containing / will currently fail (which is...

> EDIT: Sorry, I now see that the part about spaces was quite confusingly written. Spaces were just meant as an example, as they are the first thing that springs...