Re: [PATCHv6 4/4] status: better advices when splitting a commit (during rebase -i)

[Phil Hord <phil.hord@gmail.com>]

On Sun, Jun 3, 2012 at 2:30 PM, Kong Lucien <Lucien.Kong@ensimag.imag.fr> wrote:
> Add new informative help messages at the output of 'git status' when
> the user is splitting a commit. The code figures this state by
> comparing the contents of the following files in the .git/ directory:
>          - HEAD
>          - ORIG_HEAD
>          - rebase-merge/amend
>          - rebase-merge/orig-head
>
> Signed-off-by: Kong Lucien <Lucien.Kong@ensimag.imag.fr>
> Signed-off-by: Duperray Valentin <Valentin.Duperray@ensimag.imag.fr>
> Signed-off-by: Jonas Franck <Franck.Jonas@ensimag.imag.fr>
> Signed-off-by: Nguy Thomas <Thomas.Nguy@ensimag.imag.fr>
> Signed-off-by: Nguyen Huynh Khoi Nguyen <Huynh-Khoi-Nguyen.Nguyen@ensimag.imag.fr>
> Signed-off-by: Moy Matthieu <Matthieu.Moy@grenoble-inp.fr>
> ---
> The second test added by this patch fails because the case in which
> the user amend the last commit through rebase -i is not taken care of.
> We infer that the user would directly run "git commit --amend" instead
> of amending it with a rebase -i.


Maybe this is safe and logical for most workflows, or maybe this is a
convenience that makes this patch possible (I did not read the patch
yet).  But I know that I sometimes use rebase-i instead of
commit--amend because I did not realize the patch I am fixing is the
most recent one until after continuing, or because the patch I am
amending was moved into the most-recent position during the rebase-i
itself.

Just a note to consider, in case this decision was made arbitrarily.

Thanks for looking after these kinds of corner cases.

Phil