got(1) in base (was: Re: gitup(1) in base)

[Stefan Sperling]

On Tue, Feb 09, 2021 at 09:27:53PM -0000, Christian Weisgerber wrote:
> On 2021-02-08, "Osipov, Michael (LDA IT PLM)" <michael.osipov at siemens.com> wrote:
> 
> > I'd like to raise the question whether gitup(1) [1] can be included in 
> > base?
> 
> Where are we with the inclusion of got(1) [1] in base?
> 
> * It has reached a good maturity level
> * Stefan Sperling reacts quite fast to issues
> * It makes the base system complete and suitably replaces svnlite
> * It does not impose the huge complexity and dependency hierarchy of the 
> original Git client on admins and other non-developers
> 
> The original hold-up appears to have been an ex nihilo requirement
> that it must be Capsicum-ized.  Somebody was working on this, but
> has gone silent.

An intern at the FreeBSD Foundation was working on capsicum support.
This intership has now ended which is probably the reason why work on
the topic has stalled.

I am still open to working with developers from the FreeBSD community,
reviewing and integrating changes that make it easier to apply capsicum
patches to this code base.

I'd like to keep the got.git repository free of patches that won't apply
to OpenBSD because I don't want to end up having to co-maintain changes
which target other systems. I am doing this in my spare time. Maintaining
the OpenBSD version is already giving me enough things to worry about.

There is now a got-portable.git repository which currently targets Linux.
This repository is maintained by Thomas Adam.
Perhaps full FreeBSD + capsicum support could be integrated there first,
and the result integrated into FreeBSD base?
Please talk to Thomas if this sounds like a good plan.