Re: odd behavior with git-rebase

From: Neil Horman <nhorman@tuxdriver.com>
Date: Mar 26, 2012, 20:04 UTC
Message-ID: <20120326200418.GC12843@hmsreliant.think-freely.org>
In-Reply-To: <CABURp0oJwM-KtdBRVHgvOaqFVjA-MEAfJoJH=52Y=QRcgFL+3Q@mail.gmail.com>

On Mon, Mar 26, 2012 at 02:29:24PM -0400, Phil Hord wrote:

Show 21 quoted lines

> On Mon, Mar 26, 2012 at 1:12 PM, Junio C Hamano <gitster@pobox.com> wrote:
> > Neil Horman <nhorman@tuxdriver.com> writes:
> >
> >> Is there a way to differentiate a commit that is made empty as the result of a
> >> previous patch in the rebase, and a commit that is simply empty?
> >
> > An empty commit has the same tree object as its parent commit.
> >
> >> I agree, I think perhaps adding an --allow-empty option to the rebase logic, so
> >> that empty commits (or perhaps just initially empty, as opposed to commits made
> >> empty) would be very beneficial.
> >
> > Yeah, that probably may make sense.
> 
> 
> Can we have three behaviors?
> 
> A: Current mode, stop and error on empty commits
> B: --keep-empty, to retain empty commits without further notice
> C: --purge-empty, to remove empty commits without further notice
>

Yeah, I've got most of --keep-empty in a private branch here now. I was calling it allow-empty, but given (C) above, I like --keep-empty better.

I'll add --purge-empty to me todo list. and augment the rebase code to pass these options along.

One more question - The options for cherry-pick are currently mostly merged with git revert. Are there any opinions on the applicability of --keep-empty/--purge-empty to reverts?

Regards Neil