Re: [PATCH 5/5] git-rebase.txt: add a note about 'ORIG_HEAD' being overwritten

[Junio C Hamano <gitster@pobox.com>]

"Philippe Blain via GitGitGadget" <gitgitgadget@gmail.com> writes:

> From: Philippe Blain <levraiphilippeblain@gmail.com>
>
> 'ORIG_HEAD' is written at the start of the rebase, but is not guaranteed
> to still point to the original branch tip at the end of the rebase.
>
> Indeed, using other commands that write 'ORIG_HEAD' during the rebase,
> like splitting a commit using 'git reset HEAD^', will lead to 'ORIG_HEAD'
> being overwritten.

Is that a news?  If a user does "reset", the user is asking that
HEAD is changed and the old state kept in ORIG_HEAD at the same
time, so while it is not wrong per-se to say that user can clobber
ORIG_HEAD after rebase sets it first, it is pretty much expected,
no?

What would be unexpected is if "rebase" overwrote ORIG_HEAD after
user did all these other things while it gave control back and then
it took control back, and it would be worth documenting.

Having said that, I do not mind documenting this.  I am not sure "is
not guaranteed" is a good way to phrase what happens, though.

> +[NOTE]
> +`ORIG_HEAD` is not guaranteed to still point to the previous branch tip
> +at the end of the rebase if other commands that write that pseudo-ref
> +(e.g. `git reset`) are used during the rebase. The previous branch tip,
> +however, is accessible using the reflog of the current branch
> +(i.e. `@{1}`, see linkgit:gitrevisions[7]).

    `ORIG_HEAD` is set to point at the tip of the previous branch
    when `rebase` begins, but the user can run commands (e.g. "git
    reset") that overwrites `ORIG_HEAD` while `rebase` gives control
    to the user (e.g. while asking to resolve conflict).

It is excellent to mention reflog, which is very much an upward
compatible replacement of ORIG_HEAD.

Thanks.