Question about space & problem dimensions

[Lukas Korous]

Hello,

I was trying to solve problems where dim != spacedim with deal.ii and found several places in the library where I would expect spacedim and found dim instead, or found just instantiations where dim == spacedim. I would like to ask for some explanation, because I am apparently missing something.

I created a commit (on a bit old version, but it should illustrate what I mean):

- here are places where I would expect spacedim and found dim:
https://github.com/l-korous/dealii/commit/8c4fe2d47b1531b06aac22302a94551ee25db5fc#diff-1eb8ed17b743fddb50cd968dfa880a08L244

- here is a line where I would not expect that dim == spacedim is expected:
https://github.com/l-korous/dealii/commit/8c4fe2d47b1531b06aac22302a94551ee25db5fc#diff-23d9c525aded83a6d9517219b98cd46cL400

- here are some instantiations where dim == spacedim was 'hard-coded':
https://github.com/l-korous/dealii/commit/8c4fe2d47b1531b06aac22302a94551ee25db5fc#diff-6f0cbef69fc3b25c4a3ddbc3be2508dfR17

I am sure that I am missing something here, so thank you in advance for explanation.

Regards,
Lukas

[Wolfgang Bangerth]

Lukas,

All of these are in the MeshWorker part of the library for which I think only
Guido can give definite answers. My suspicion is that these parts were written
before we had the ability to solve problems with dim != spacedim and were
never converted.

The patches that convert things like
Triangulation<dim> -> Triangulation<dim,spacedim>
DoFHandler<dim> -> DoFHandler<dim,spacedim>
are obviously correct and if you'd like to propose an incremental pull request
that already gets this part out of the way, I would be happy to approve this.

I *suspect* that most of the other changes are correct as well, but let Guido
confirm.

Best
Wolfgang


--
------------------------------------------------------------------------
Wolfgang Bangerth email: bang...@math.tamu.edu
www: http://www.math.tamu.edu/~bangerth/

[Lukas Korous]

OK, thanks for the explanation. There is some more work needed for support of completely general dim, spacedim, e.g. because of what is described in this commit:

https://github.com/l-korous/dealii/commit/f4aaea8f6380d8c1d02838bf3078a53b75e19fb7

Lukáš Korous

--
The deal.II project is located at http://www.dealii.org/
For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en
--- You received this message because you are subscribed to a topic in the Google Groups "deal.II User Group" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/dealii/rQaidH4e928/unsubscribe.
To unsubscribe from this group and all its topics, send an email to dealii+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Wolfgang Bangerth]

On 08/18/2015 06:59 AM, Lukas Korous wrote:
> OK, thanks for the explanation. There is some more work needed for support of
> completely general dim, spacedim, e.g. because of what is described in this
> commit:
>
> https://github.com/l-korous/dealii/commit/f4aaea8f6380d8c1d02838bf3078a53b75e19fb7

But this change also makes sense. 'dim' always has to be <= 'spacedim'.

Best
W.

[Lukas Korous]

Hi Wolfgang, I am missing something then. I thought that 'dim' is (e.g. in the case of DoFHandler) the number of equations. Then

'dim' always has to be <= 'spacedim' would not be true for any larger system (e.g. Euler equations).

Regards,
Lukas

[Wolfgang Bangerth]

On 08/18/2015 10:26 AM, Lukas Korous wrote:
> Hi Wolfgang, I am missing something then. I thought that 'dim' is (e.g. in the
> case of DoFHandler) the number of equations. Then

No, dim is the dimensionality of the cells. For quadrilaterals, dim=2. For
hexahedra, dim=3. spacedim is the dimension of the space in which they live.
So, for example, if you have dim=2, spacedim=3, then you have a mesh that
meshes a surface.

Clearly you can't have a mesh of quadrilaterals that live in 1d, though.

(Or are you referring to something specific to MeshWorker that I don't
understand?)

[Lukas Korous]

Sure - you are right, I was thinking about something else.

OK - I will try to create a pull request from some of the relevant changes.

Lukáš Korous

                                www: http://www.math.tamu.edu/~bangerth/