[Dune] Interface of PersistentContainer

Martin Nolte nolte at mathematik.uni-freiburg.de
Wed Feb 20 18:00:16 CET 2013


Hi Christian,

sorry for the late answer. I just committed a documentation of the interface 
for a persistent container (including a slight overhaul, hopefully making it 
easier to understand).

Strictly speaking, revision 8871 could be considered an interface change, but 
I currently don't consider the persistent container part of the official 
interface. Anyway, all implementations ship deprecated versions of the old 
methods. If someone objects to this patch, I would ask him to simply revert 
it. We can then discuss on FlySpray.

I hope the documentation answers most of your questions. As soon as I find the 
time, I will add a specialization of some persistent containers for 
AlbertaGrid (using their internally managed data vectors), to provide yet 
another implementation.

Again, sorry for the delay. Proper documentation can really take some time...

Best,

Martin

On 02/05/2013 02:34 PM, Christian Engwer wrote:
>
> Dear all,
>
> I wanted to implement the PersistentContainer for YaspGrid and
> stumbled upon some questions/issues regarding the interface.
>
> a) The "official" implementation has a method "size()". What is the
>     idea for this method? As it is not possible to access the data in a
>     vector like style, I don't understand the purpose. Depending on the
>     data storage stragtegy it can be quite expensive to obtain this
>     information.
> b) what is the desired semantic if I try to access an entry which is
>     not stored? E.g. I store the old soluation and access cell-entries
>     after a refine, without adjusting the size of the container?
>
> Christian
>
>
> _______________________________________________
> Dune mailing list
> Dune at dune-project.org
> http://lists.dune-project.org/mailman/listinfo/dune
>

-- 
Dr. Martin Nolte <nolte at mathematik.uni-freiburg.de>

Universität Freiburg                                   phone: +49-761-203-5630
Abteilung für angewandte Mathematik                    fax:   +49-761-203-5632
Hermann-Herder-Straße 10
79104 Freiburg, Germany




More information about the Dune mailing list