Re: Regression in 'git branch -m'?

[Jeff King <peff@peff.net>]

On Thu, Oct 05, 2017 at 07:25:52PM +0200, Andreas Krey wrote:

> I got something that looks like a regression somewhere since 2.11.
> This script
> 
>   set -xe
>   rm -rf repo
>   git init repo
>   cd repo
>   git commit -m nix --allow-empty
>   git branch -m master/master
>   git rev-parse HEAD
>   git branch
>   git status
> 
> causes .git/HEAD to still contain 'ref: refs/heads/master' and to fail
> in the rev-parse step with
> 
>   + git rev-parse HEAD
>   HEAD
>   fatal: ambiguous argument 'HEAD': unknown revision or path not in the working tree.
>   Use '--' to separate paths from revisions, like this:
>   'git <command> [<revision>...] -- [<file>...]'
> 
> This is with 2.15.0.rc0; with 2.11.0 (and 2.11.0.356.gffac48d09) it still works.

So this turned out to be quite an interesting bug to explore. I think
the solution I ended up with in the second patch is the right thing. I'm
adding Michael to the cc for wisdom on the ref code, though I think the
bug I'm fixing actually goes back to the early days of Git.

Earlier I blamed Duy's 31824d180d. And that is the start of the
regression in v2.15, but only because it fixed another bug which was
papering over the one I'm fixing here. :)

  [v1 1/2]: t3308: create a real ref directory/file conflict
  [v1 2/2]: refs_resolve_ref_unsafe: handle d/f conflicts for writes

 refs.c                  | 15 ++++++++++++++-
 t/t1401-symbolic-ref.sh | 26 +++++++++++++++++++++++++-
 t/t3200-branch.sh       | 10 ++++++++++
 t/t3308-notes-merge.sh  |  2 +-
 4 files changed, 50 insertions(+), 3 deletions(-)

-Peff