Re: [PATCH 0/5] Fast forward strategies allow, never, and only

From: Sverre Hvammen Johansen <hvammen@gmail.com>
Date: Apr 24, 2008, 05:39 UTC
Message-ID: <402c10cd0804232239p3e49a6d0vdc1ff9acb3636a17@mail.gmail.com>
In-Reply-To: <7vhcdu2uu5.fsf@gitster.siamese.dyndns.org>

On Tue, Apr 22, 2008 at 12:48 AM, Junio C Hamano <gitster@pobox.com> wrote:

>  Sorry, but I am with a rather big backlog and am reluctant to go back the
>  archive a looong way to pick up and comment on a series when not many
>  people are wondering what happened to the wonderful series ;-)  Care to
>  resend and ask for comments from people?

 I resending these patches for you to coment.  The patch series
consists of the following five patches:

  0001-New-merge-tests.patch
  0002-Introduce-ff-fast-forward-option.patch
  0003-Restructure-git-merge.sh.patch
  0004-Head-reduction-before-selecting-merge-strategy.patch
  0005-Introduce-fast-forward-option-only.patch

The first patch add some tests. The second, fourth, and fifth adds new features and they are all trivial. I was able to make the fourth patch trivial as well by actually doing the real work of finding the reduced parents in the third patch. The third patch computes the reduced parents but uses it only to determine whether we are up-to-date or do a fast forward.

There are probably some minor adjustments to the documentation we should do. The patch series uses the term actual-parents and reduced-parents in the code and the documentation. Maybe we should use the term actual-heads and reduced-heads instead? I am not sure that all the documentation for 0004 should be included. Please give me some advise regarding this.

-- 
Sverre Hvammen Johansen