thanks. That little bit at the bottom must be a cut/paste, earlier version of
the note, not updated :-)...
Given that the master is for the 2.5.x series, what is the philosophy / naming
convention for working on the 3.x series?
-Marshall
On 7/25/2019 5:39 PM, Richard Eckart de Castilho wrote:
> Hi,
>
>> On 25. Jul 2019, at 20:56, Marshall Schor <msa@schor.com> wrote:
>>
>> I found the writeup for how people should work with uimaFIT on github in
>> .github/CONTRIBUTING.md.
>>
>> (it might be better to put this at the root. GitHub docs say it can be at the
>> root, in docs/ or .github/, but people scanning the project to see how to
>> contribute might not look inside .github...)
> Sure, we can move it.
>
>> I see there's branches:
>> 2.0.x
>> 2.1.x
>> 3.0.x
>> master
>> release/2.5.0
>> release/3.0.0
>> release/3.0.0-beta-1
>>
>> Can you say what each of these mean, and if "master" is current used? (It seems
>> at some back level, perhaps 2.4.0).
> "master" is the branch against which people usually would want to file their pull requests.
Currently, this is the 2.5.x series.
>
> There are maintenance branches for some older series (2.0.x, 2.1.x) as well as for the
v3 code (3.0.x series).
>
> The "release/2.5.0" branch I am using to prepare the 2.5.0 release. Once the release
is out, it gets merged into master. Probably I will switch master to the v3 code after this
release (by merging 3.0.x into master and then deleting 3.0.x).
>
>> When you do a release, is it from a 3.0.x (etc) branch?
> I would usually ask INFRA to protect maintenance and main development branches (master,
2.*, 3.*, etc) and configure them such that nobody can directly write to them. Which means
that I am using branches such as release/2.5.0 and release/3.0.0 to prepare release and them
use a PR to merge them into their respective target branches (master, 3.0.x) after the release.
The release/3.0.0 and release/3.0.0-beta-1 seem to be leftovers that I forgot to delete.
>
>> Looking at the CONTRIBUTING.md in the .github folder, it says
> Hm?
>
> Cheers,
>
> -- Richard
>
>
>
|