search

radical-cybertools / radical.ensemblemd.mdkernels

Other
0 stars 0 forks source link

confusing branching #2

Open marksantcroos opened 9 years ago

marksantcroos commented 9 years ago

Can we stick to conventions regarding branching?

oleweidner commented 9 years ago

Sure.

oleweidner commented 9 years ago

Branches are in sync again. I will keep "release" in place as it might be used in some documentation / installation instructions.

I have also created a 'regular' devel branch.

marksantcroos commented 9 years ago

I will keep "release" in place as it might be used in some documentation / installation instructions.

Aren't we early enough in the life of this project to just fix the documentation?

andre-merzky commented 9 years ago

An, iff we keep it, could we use the same branch names as for the other projects? ;) I am sure you expected this comment, right? :D

oleweidner commented 9 years ago

There's some testing for ExTASY 0.1 going on at the moment. Removing the "release" branch would be disruptive. I'll remove it before ExTASY 0.2.

marksantcroos commented 9 years ago

On 25 Sep 2014, at 12:09 , Ole Weidner notifications@github.com wrote:

There's some testing for ExTASY going on at the moment. Removing the "release" branch would be disruptive. I'll remove it before ExTASY 0.2.

Fair enough, and thanks for merging!

marksantcroos commented 9 years ago

Devel and release are in addition to being behind master for some number of commits also one commit ahead of master, but I think master is being used?

I believe we wanted to get rid of release?

Is this thing on pypi now or still released from source? I could find an explicit reference to install it in the ExTASY documentation but not with EnMDTK.

vivek-bala commented 9 years ago

I'm not sure about the other projects, but ExTASY uses mdkernels/master. It's not on pypi as well.

EnMDTK does not use mdkernels.

marksantcroos commented 9 years ago

I'm not sure about the other projects, but ExTASY uses mdkernels/master. It's not on pypi as well.

Thanks!

EnMDTK does not use mdkernels.

Ok, this surprised me, but I see indeed in the code now that EnMDTK uses its own kernels.

Oh well, more ammunition for https://github.com/radical-cybertools/radical.pilot/issues/498 :-)

oleweidner commented 9 years ago

Indeed -- merging and outsourcing of the application kernels might be a nice (semester?) project for the future, especially since there are more than one stakeholder in the Kernels now.

Any takers? ;-)