Some of the interfaces were designed such that they are in agreement
with the porous medium flow framework where we make them work for
both cell-centered as well as the box scheme. In the geomechanics
framework we now make use of the fact that cell-centered schemes are not
considered and use more FEM-like interfaces for the parameters required
for stress evaluations. This might enable us to rename/redefine these
interfaces to work with both FEM as well as Box. However, if
cell-centered schemes for geomechanics are to be implemented one day,
the whole structure would have to be overthought.

We currently dont support them because they have become too old.
After refactoring they will be reincluded in a future release

Kilian Weishaupt authored 7 years ago and Timo Koch committed 7 years ago

* Dune > 2.5 uses a new interface and only required GridView as template
  argument for the mapper type
* for the constructor, an additional argument is now required:
  elementMapper_(gridView, Dune::mcmgElementLayout())
  vertexMapper_(gridView, Dune::mcmgVertexLayout())

(cherry picked from commit 6b4dc071)

* Dune > 2.5 uses a new interface and only required GridView as template
  argument for the mapper type
* for the constructor, an additional argument is now required:
  elementMapper_(gridView, Dune::mcmgElementLayout())
  vertexMapper_(gridView, Dune::mcmgVertexLayout())

Sina Ackermann authored 8 years ago and Timo Koch committed 7 years ago

(cherry picked from commit 0410f946)

(cherry picked from commit 3b62b6af)

[test] Fix el2p problem

See merge request !417

* 02e54008 erroneously
  replaced Yasp with AluGrid

Feature/pieceofcake

See merge request !354

Replacing LinearSpacer/PowerLawSpacer structure with the grading option
from gridcreator.hh, which allows to define multiple sets of increasing/
decreasing point distances for angular, radial and axial coordinates.

Beatrix Becker authored 8 years ago and Timo Koch committed 8 years ago

(cherry picked from commit 286ba6c7)

The global object can create an empty temporary local object that
can be bound to an element/stencil. This is closer to the
current dumux-stable implementation and still allows to switch
between global caching (for smaller problems where memory is not an issue)
and local caching (variables are precomputed for each element assemble temporarily)

Works for cc and box (in 2d). Box needs some performance improvement,
probably when creating the box geometries. But also other parts
might be possbile to optimize.

Dennis Gläser authored 8 years ago and Timo Koch committed 8 years ago

This commit introduces the necessary changes and new classes for a
cctpfa elastic model. Checking the results suggest, that a simple
tpfa scheme is not suitable for linear elasticity, not even on structured
grids. It is thus questionable wether or not this implementation will be deleted
in the future. Note that it only works in 2d so far.

Everything in our .hh files is defined within the namespace `Dumux`.
There were numerous occurrences of `Dumux::` qualifiers that were
unnecessary. Remove those occurrences. Necessary qualifications
remain, such as `typedef Dumux::H2O<Scalar> H2O;`.

this means
* the first vtk (normally the initial solution) is *-00000.vtu
* if all timesteps are writen the time step number and vtk number correspond

Move all testing related scripts to subfolder testing

Kilian Weishaupt authored 9 years ago and Timo Koch committed 9 years ago

*remove #include <dune/grid/io/file/dgfparser/dgfyasp.hh>
*remove #include <dune/grid/io/file/dgfparser.hh>
*ect

* liquid phase or gas phase can by chosen via the property system, so this fluidsystem is superfluous

* Remove empty constructor
* Remove wrongly included config.h which is only needed in cc files
* Remove else after return in if branch