This type of change never gets old...

It is wrong with GCC 5.x anyway.

Remove use of isParallel.

canCommunicate is already part of Dune 2.4.

See merge request !27

canCommunicate is already part of Dune 2.4.

For binary compatibility, Dune only uses the C bindings
of MPI. MPI_FOUND is deprecated from CMake but Dune does
defines it and encourage its use.
(reviewed by timok)

[dumux, test] fix shadowing of variables

Fix all warnings triggered with -Wshadow.

See merge request !25

Fix all warnings triggered with -Wshadow.

[el2p] fix the model for two-dimensional settings

At various places, the code still assumed 3d. Generalize it by
replacing [2] by [dimWorld-1] or by appropriate if statements.

See merge request !24

Kilian Weishaupt authored Nov 13, 2015 and Bernd Flemisch committed Nov 18, 2015

It is now possible to implement a model specific
onBoundary(const int dofIdxGlobal) function.

At various places, the code still assumed 3d. Generalize it by
replacing [2] by [dimWorld-1] or by appropriate if statements.

(reviewed by martinb)

[el2p] fix the El2PAMGBackend for sequential grids and HAVE_MPI==1

The fix for the general AMGBackend in commit 21c2b3d8 didn't reach
the El2PAMGBackend, since the el2p model defines its own AmgTraits.
Resolve the issue by using template specialization and
Dune::Capabilities.

See merge request !23

The fix for the general AMGBackend in commit 21c2b3d8 didn't reach
the El2PAMGBackend, since the el2p model defines its own AmgTraits.
Resolve the issue by using template specialization and
Dune::Capabilities.

[gridcreator] clear off bug, std::string instead of int for reading refinement type and closure type from input file

The old code only worked because the major version changed.

The test magic disappeared in dune3.0+

The dumux test macro doesn't need to add flags anymore.

Fixes memory leaks if a fluidsystem with more than 4 components
is used in a 2D case.
The problem was in each loop the variables i and j for the SCV
corners were shadowed by the component indices.

Signed-off-by: Christoph Grüninger <christoph.grueninger@iws.uni-stuttgart.de>

Most of the properties tags are unneccesary, as they are already defined in 1p2cproperties or 2pproperties, respectively.

(reviewed by martins)

Fix/generallensproblem

Use the parameter macro instead of the tree directly.

See merge request !19

Since ElOnePTwoC doesn't inherit from OnePTwoC, the FluidState has
to be set explicitly.

Reviewed by Natalie.

[implicit] Let models only modify/set their own TypeTag

The defaults modified properties of Parent TypeTag like BoxModel
and NonIsothermal. This leads to conflicts if one uses more than one
model in a problem. Now all models only set their respective TypeTags
and property defaults.

See merge request !22

TwoPMinc properties are set for TwoPMinc TypeTag 
instead of BoxTwoPMinc.

This is changed for the ForchheimerCoeff

The defaults modified properties of Parent TypeTag like BoxModel
and NonIsothermal. This leads to conflicts if one uses more than one
model in a problem. Now all models only set their respective TypeTags
and property defaults.

[cleanup] Remove all leftover occurences of deprecated map()

Replace by the index/subIndex method

See merge request !21

Replace by the index/subIndex method

Remove outdated INSTALL, point to online resources.

There is no point in keeping one more outdated installation guide.

See merge request !20

It was not clear where the name of the test is coming from, because
it was set in the cc file. Deduce the name from the modeltype now.

The function returning the partial pressure was accidentally removed
from the immiscible fluidstate. This caused an compiler error, e.g., when
the water-air fluidsystem is used in 2p models.

Reviewed by Thomas

Return an std::string. Don't use the parameter tree directly.
Parameters are registered when using the macro. Thus our summary
output with unused parameters only works when using our macro.

Line breaks are not allowed in cmake strings. After
version 3.0 linebreaking with backslash is possible.