Open Stack

On Mon, Sep 22, 2014 at 7:51 AM, Jay S. Bryant
<jsbryant at electronicjungle.net> wrote:
>
> On 09/21/2014 07:37 PM, Matt Riedemann wrote:
>>
>> When I'm essentially +2 on a change but for a small issue like typos in
>> the commit message, the need for a note in the code or a test (or change to
>> a test), I've been doing those myself lately and then will give the +2.  If
>> the change already has a +2 and I'd be +W but for said things, I'm more
>> inclined lately to approve and then push a dependent patch on top of it with
>> the changes to keep things from stalling.
>>
>> This might be a change in my workflow just because we're late in the
>> release and want good bug fixes getting into the release candidates, it
>> could be because of the weekly tirade of how the project is going down the
>> toilet and we don't get enough things reviewed/approved, I'm not sure, but
>> my point is I agree with making it socially acceptable to rewrite the commit
>> message as part of the review.
>
> Matt,
>
> This is consistent with what I have been doing for Cinder as well. I know
> there are some people who prefer I not touch the commit messages and I
> respect those requests, but otherwise I make changes to keep the process
> moving.
>
> Jay

This is also consistent with how I've been doing things in Ironic and
I have been encouraging the core team to use their judgement when
doing this as well -- especially when it's a patch from someone we
know won't get back to it for a while (eg, because they're on
vacation) or someone that has already OK'd this workflow (eg, other
members of the core team, and regular developers we know).

-Devananda