Re: [PATCHv4 2/2] pull --rebase: Avoid spurious conflicts and reapplying unnecessary patches
- From
- Santi Béjar <santi@agolina.net>
- Date
- Aug 12, 2010, 22:51 UTC
- Message-ID
- <AANLkTinJK3YZ0sojnpiH0KZvFFC38X8LFr3U1ZrU_UXG@mail.gmail.com>
- In-Reply-To
- <AANLkTinqW4h-SG21ZEAvhKEPJ46WfBRyYboUsfXjy7Zn@mail.gmail.com>
On Thu, Aug 12, 2010 at 11:02 PM, Elijah Newren <newren@gmail.com> wrote:
Show 21 quoted lines
> Hi,
>
> On Thu, Aug 12, 2010 at 8:37 AM, Santi Béjar <santi@agolina.net> wrote:
>> diff --git c/git-pull.sh w/git-pull.sh
>> index a09a44e..c1617d5 100755
>> --- c/git-pull.sh
>> +++ w/git-pull.sh
>> @@ -214,7 +214,10 @@ test true = "$rebase" && {
>> do
>> if test "$reflog" = "$(git merge-base $reflog $curr_branch)"
>> then
>> - oldremoteref="$reflog"
>> + if test "$reflog" != $(git merge-base $reflog
>> $remoteref)
>> + then
>> + oldremoteref="$reflog"
>> + fi
>
> How does this help? I've been trying to scratch my head trying to
> figure out a case where it could affect the outcome, and am struggling
> to come up with one.Me too :)
The only case where it makes a difference, it is wrong. I was trying to be extra cautious, but ... too much in this case.
Show 17 quoted lines
> >> @@ -273,6 +276,14 @@ then >> exit >> fi >> >> +if test true = "$rebase" >> +then >> + if test "$oldremoteref" = $(git merge-base $oldremoteref $merge_head) >> + then >> + unset oldremoteref >> + fi >> +fi >> + > > This was indeed my original patch, then I just (incorrectly, as you > pointed out) moved other bits of code to be with this so that the > determination of oldremoteref was all in one place.
OK. Then for your original patch:
Acked-by: Santi Béjar <santi@agolina.net>
Thanks, Santi