dolfin team mailing list archive

["Garth N. Wells" <g.n.wells@xxxxxxxxxx>]

Anders Logg wrote:
> On Thu, Oct 26, 2006 at 01:16:27PM +0200, Johan Jansson wrote:
>> We had an idea to create a branch of DOLFIN where we could work on
>> (and break) the mesh while still keeping a more stable DOLFIN branch
>> with the old mesh. Now it turned out the other way around, the mesh
>> was replaced in the main DOLFIN branch while the new branch was kept
>> stable. Anyway, I think it has worked rather well to have two
>> branches, high-level development has been able to continue in the
>> stable branch, and kernel development has been done in the main
>> development branch.
>>
>> Notifications from the stable branch (called "dolfin-kth") have not
>> been sent to this list though. I've turned that on now. There have
>> only been a few changesets, you can view them here:
>>
>> http://www.fenics.org/hg/dolfin-kth
>>
>> Please comment if you have ideas how to organize the branches, if we
>> should have two branches inside the same repository (doesn't work so
>> smoothly with Mercurial in my experience), or whether we should have
>> separate repositories (even more than two perhaps).
>>
>> Starting/cloning a repository is so simple in Mercurial (everyone has
>> a private repository anyway when they develop), so I think it's
>> natural that every repository represents a branch.
>>
>>   Johan
>> _______________________________________________
>> DOLFIN-dev mailing list
>> DOLFIN-dev@xxxxxxxxxx
>> http://www.fenics.org/mailman/listinfo/dolfin-dev
> 
> I'm not so sure I like the branched development we have now with a
> separate dolfin-kth. The consequence of this development is that the
> core development takes place in the main dolfin tree (mostly by Garth
> and myself), while you work on your own things in dolfin-kth.
>

Agree.

> If you instead worked in the main dolfin tree, you would be forced to
> keep up with and contribute to the development of the core
> functionality which I would like better.
> 

Agree.

> I think separate trees should be something that we create temporarily
> for testing and implementing new features before merging back into the
> main tree. It should not be something permanent since it would then in
> practice be a fork.
> 

This I really agree on. I'm all for temporary branches when making major
changes. A problem when making a major transition is that development on
other aspects halts because the code doesn't compile or doesn't yet work
properly. A big test will be the new build system. It will naturally
take some time to get it fully functional, but it shouldn't stall other
developments. Could be a good chance to test a temporary branch.

Garth

> /Anders
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev
>