Re: About the Git branching model

2018-02-02 Thread Dawn Shepherd
Hi Animesh,

The following lines are my idea.

The version number could be in a popular format
major.minor.patch

1. The master branch keeps the major & minor SNAPSHOT versions. (e.g. 
1.0.0-SNAPSHOT, 1.1.0-SNAPSHOT, 2.0.0-SNAPSHOT)
2. The release branch (e.g. branch-1.0) keeps the patched SNAPSHOT versions. 
(e.g. 1.0.1-SNAPSHOT, 1.1.2-SNAPSHOT)
3. Use git tags (e.g. v1.0.3) to cut a release. Once released, the SNAPSHOT 
suffix needs to be removed. And then, the release branch moves on to the next 
patched SNAPSHOT version.

Maybe my idea is the same as yours? 

> On 31 Jan 2018, at 02:33, Animesh Trivedi  wrote:
> 
> Hi Dawn,
> 
> what specifically do you have in mind? I agree that once we start to have
> periodic releases (hopefully soon) we will tag corresponding git branch
> with the version number. Since, there is no release, all the code is in the
> master branch.
> 
> If you tell us what do you have in mind, we can include that structure :)
> 
> Thanks,
> --
> Animesh
> 
> On Tue, Jan 30, 2018 at 3:36 PM, Dawn Shepherd 
> wrote:
> 
>> Hi all,
>>I am not in the Crail team and I have a suggestion. Perhaps the
>> Crail project could use Spark's branching model? I think it is very simple
>> and practical



Re: About the Git branching model

2018-01-30 Thread Animesh Trivedi
Hi Dawn,

what specifically do you have in mind? I agree that once we start to have
periodic releases (hopefully soon) we will tag corresponding git branch
with the version number. Since, there is no release, all the code is in the
master branch.

If you tell us what do you have in mind, we can include that structure :)

Thanks,
--
Animesh

On Tue, Jan 30, 2018 at 3:36 PM, Dawn Shepherd 
wrote:

> Hi all,
> I am not in the Crail team and I have a suggestion. Perhaps the
> Crail project could use Spark's branching model? I think it is very simple
> and practical


About the Git branching model

2018-01-30 Thread Dawn Shepherd
Hi all,
I am not in the Crail team and I have a suggestion. Perhaps the Crail 
project could use Spark's branching model? I think it is very simple and 
practical