Didier 'OdyX' Raboud writes ("Bug#972098: dgit assumes 'master' as main
branch"):
> Le mardi, 13 octobre 2020, 00.23:12 h CEST Ian Jackson a écrit :
> > Didier 'OdyX' Raboud writes:
> > > on various packaging repositories I routinely use dgit on, it has started
> > > to fail on me with the following error:
> > This is odd. dgit doesn't normally care much about branches.
>
> It surprised me too
I can't repro this. Indeed I just checked and the "quilt" test case
in the dgit test suite uses a git tree with no master branch. Most
puzzling.
> > Can you please run the command again with (let's say) -DD ?
>
> Hereby attached.
Thanks. I think the failing command is this
runcmd @git, qw(checkout -q), (qw(master dgit-view)[do_split_brain()]);
which runs in the playtree - ie, a little private space of dgit's own,
with a different refs namespace to your tree. do_split_brain() is
false which seems right. I don't see where the playtree master branch
comes from, even on my working test case.
I am out of time now. I will debug this more later.
Ian.
--
Ian Jackson <ijack...@chiark.greenend.org.uk> These opinions are my own.
Pronouns: they/he. If I emailed you from @fyvzl.net or @evade.org.uk,
that is a private address which bypasses my fierce spamfilter.
