poor git style in ACL patches

[Richard Sharpe]

On Wed, May 16, 2012 at 11:36 PM, Andrew Bartlett <abartlet at samba.org> wrote:
> On Wed, 2012-05-16 at 22:00 -0700, Jeremy Allison wrote:
>> On Thu, May 17, 2012 at 12:55:59AM -0400, simo wrote:
>> > On Thu, 2012-05-17 at 05:09 +0200, Andrew Bartlett wrote:
>> > > commit 367a644c4d91531faf8b2ce9a167fc196da12422
>> > > Author: Andrew Bartlett <abartlet at samba.org>
>> > > Date:   Mon May 14 17:11:09 2012 -0700
>> > >
>> > >     We need to split things up into a new helper function
>> > > add_current_ace_to_acl() in order for there to be more posix ACL
>> > > elements than NT ACL elements (so a group SID can own a file, but also
>> > > get the group permissions that will be honoured)
>> >
>> > Andrew,
>> > it would be really appreciate if you could stick to git standard of
>> > providing a short 'subject' in the first line, then leave one empty, and
>> > then put the commit message wrapped to 72 chars.
>> >
>> > This is also the standard used in the kernel and makes logs a lot more
>> > readable (also in gitk).
>>
>> This one might be my fault, as I hand-patched some of
>> Andrews changes, made the edits I wanted and then commited
>> them under his identity with my sign-off (as they were 99.999%
>> his changes).
>>
>> I may have missed the first line of his original commit
>> message when doing the commit, so this might be my error
>> in the review.
>
> Using git rebase -i and git commit --amend is a good way to keep the
> original commit detail intact, while making small changes.

Yes, I routinely use git rebase -i to rearrange commits or squash
related commits into one single commit so that people do not have to
deal with intermediate commits that might also break things.

It is very powerful and worth getting to know.

-- 
Regards,
Richard Sharpe
(何以解憂？唯有杜康。--曹操)