Edward Thomson

April fools? No, but seriously, that's a lot slower. I don't know why that might be the case offhand (I don't use libgit2's blame, I don't think that anybody's actually...

I don't think that will help - blame/annotate is a well solved problem, and jgit's internals are sufficiently different that it's not a straight port. Even taking git's blame code...

Is this philosophical or practical? What I mean is - are you relying on very old ciphers? Can you use the options functionality that we have to control that? We...

@Pyenb What do you _want_ hook support to look like? Are these the APIs that you want?

What's amazing about this is that `safe.directory` is case _sensitive_ on macOS, so this is not actually honoring `core.ignorecase`. `safe.directory` is an unmitigated disaster.

Hi @flagarde - thanks for the details (and for your patience with my limited understanding of cmake, and my hurried PR review, where I still left TODO comments for myself...

I think that it's okay to keep it as "changes requested" - that makes sense within the context of what we're doing (we found a vulnerability, we don't want you...

I added a suggested change to bump to libgit2 v1.8.1. If that doesn't unblock us, I've prepared a v1.8.2 rc1 that reverts a change to `git_commit` functions. I'm not sure...

👋 I'm sad to see this PR is so involved. Can you help me understand what the problem(s) is (are) and how I can help? Do we need to drag...

Thanks for the details. These are often things that seem not so obvious in the small scale. It's only when you see them in the macro that you realize just...