Bernd Flemisch authored 8 years ago and Kilian Weishaupt committed 8 years ago

This concerns comments, output and local variable names.
Also fix some comments where liquid has been used instead of gas.

(cherry picked from commit cc2f9ee2)

Kilian Weishaupt authored 8 years ago and Timo Koch committed 8 years ago

* Use ADL for std:: functions
* Adapt doxygen docu

Kilian Weishaupt authored 8 years ago and Timo Koch committed 8 years ago

* Adapt to new interface
* Fix error in neumann term (use wPhase density instead of gPhase density)
* Adapt ref files accordingly

Kilian Weishaupt authored 8 years ago and Timo Koch committed 8 years ago

* Adapt to new interface
* Rename variables in ref solution and CMakeList.txt
* Use \ to escape the brackets in CMakeList.txt

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

(cherry picked from commit 134c8aa2)

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

(cherry picked from commit 179b0a10)

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

(cherry picked from commit 286ba6c7)

The tracer model solves transport equations for multiple tracers when
a velocity field is given by the user (possibly by a 1p simulation).

Note: tests still fail

Note: the box tests still fail

This commit also ports the 1pniconduction test to the new output module

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

- The laws store a pointer to the spatial parameters instead of an initial field (more user-friendly).
- The permeability type has to be stated publicly in the spatial parameters

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

The box test passes here, but strangely the cc test fails

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

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.[1p] port ni tests to new interface

The tests fail, but they do so because the output has changed. This has
to be checked again.

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

Unfortunately, the box 1p test does not pass. It remains to check why and
if it didn't do so before as well.

Fix/restart and episodes

For me it would be more intuitive to first advance the time level and then do all the processing and file writing. However this is more complicated, therefore I shifted the restart stuff to the same level as the vtk output stuff.

See merge request !310

[sequential grid adaptation] Fix problems due to commit 42dd5278

Set GridAdaptModel with second template argument (true/false for adaptivity) depending on AdaptiveGrid property. Include file for default initialization indicator. Clean-up.

See merge request !313

Christoph Grüninger authored 8 years ago and Thomas Fetzer committed 8 years ago

Speed-up multidomain tests

[tests][multidomain] Use tabulated H2OAir FluidSystem
* speeds-up test by 5
* hopefully makes test more robust when comparing to reference solution

See merge request !309

* Set the property for ImplicitBase and define default of 1.0.
* Remove any obsolete declarations from all ported models