dumux-repositories issueshttps://git.iws.uni-stuttgart.de/groups/dumux-repositories/-/issues2023-12-08T21:19:22Zhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1321Brinkman helper for CVFE2023-12-08T21:19:22ZTimo Kochtimokoch@math.uio.noBrinkman helper for CVFE!3715 adds a helper function for a Brinkman term in the Navier-Stokes model using staggered FV. It should be easy to also provide the CVFE implementation (simply condition on the discretization method in the helper function).
The test wo...!3715 adds a helper function for a Brinkman term in the Navier-Stokes model using staggered FV. It should be easy to also provide the CVFE implementation (simply condition on the discretization method in the helper function).
The test would need to be extended then.3.9https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1199make a detailed description of installation of dumux on MacOS2023-12-13T11:13:10ZDavid Wernermake a detailed description of installation of dumux on MacOSThere are three major os-platforms which users use mostly, one of them MacOS.
With a few tweaks one can on all three be as productive develop DuMux as on Linux, at least
if one has recent os-release. That's why the project should provid...There are three major os-platforms which users use mostly, one of them MacOS.
With a few tweaks one can on all three be as productive develop DuMux as on Linux, at least
if one has recent os-release. That's why the project should provide easy applicable descriptions
for all three of them on the website and this not to deep hidden in it.
As far as I know one would install [homebrew or shortly named brew](https://brew.sh/) there are packages from the MacOS-software selection tool (shop) and there are packages from [macports](https://ports.macports.org/).
With brew easily install the necessary packages but it may also interfere with already selected tools.
I tried to document the possible package names on the [DuMux wiki](https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/wikis/Package-Names,-Installation-of-DuMux-on-various-systems).
For the compiler toolchain, I see several options: Clang from the system, Clang from updated Xcode, GCC from brew (likely together with update Xcode), Clang from brew, or also clang or gcc from macports. Likely some ways are more preferable. So give hints. Provide if necessary an adapted cmake.opts of your solution to the user.https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1188Update all grid caching classes with multithreaded loops2022-09-21T22:18:45ZTimo Kochtimokoch@math.uio.noUpdate all grid caching classes with multithreaded loopsUpdate all grid caching classes with multithreaded loopsUpdate all grid caching classes with multithreaded loops3.6Timo Kochtimokoch@math.uio.noTimo Kochtimokoch@math.uio.nohttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1180New Compositional Staggered2023-10-26T08:13:30ZNed ColtmanNew Compositional Staggeredmentioned in #1115
Implementation and ported tests addressed in !2986 .
New Analytical solution and convergence test implemented in !3044 . Also implemented for old staggered in !3561 .
Troubleshooting branch: `feature/misc_naviersto...mentioned in #1115
Implementation and ported tests addressed in !2986 .
New Analytical solution and convergence test implemented in !3044 . Also implemented for old staggered in !3561 .
Troubleshooting branch: `feature/misc_navierstokesnc` .
Using this issue to collect all of the ideas and progress we have made (the MRs are not organized for this)3.8Ned ColtmanNed Coltmanhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1177More flexible `FVElementGeometry` interface2023-12-13T11:13:10ZDennis GläserMore flexible `FVElementGeometry` interfaceIt may be better to pull out some functions of the `SubControlVolume` and `SubControlVolumeFace` classes into the `FVElementGeometry`. This can significantly reduce the memory overhead (see #1173 and !3183).
For instance, we could do
`...It may be better to pull out some functions of the `SubControlVolume` and `SubControlVolumeFace` classes into the `FVElementGeometry`. This can significantly reduce the memory overhead (see #1173 and !3183).
For instance, we could do
```cpp
const auto outsideScv = fvGeometry.outsideScv(scvf);
```
instead of
```cpp
const auto outsideScv = fvGeometry.scv(scvf.outsideScvIdx());
```
This way, users don't need to know that there is an index stored and it also doesn't force us to actually store an index (which is probably not necessary for box, for instance). It also forces us to write more consistent code between box and cc, because we sometimes do
```cpp
// we know that for cc the inside index is the dof index
const auto insideDofIndex = scvf.insideScvIdx();
```
although a version that is compatible with box schemes would be
```cpp
const auto insideDofIndex = fvGeometry.scv(scvf.insideScvIdx()).dofIndex();
```https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1174Use spell checker for code2022-07-28T12:31:26ZTimo Kochtimokoch@math.uio.noUse spell checker for codehttps://github.com/codespell-project/codespell
see https://gitlab.dune-project.org/core/dune-common/-/merge_requests/1149https://github.com/codespell-project/codespell
see https://gitlab.dune-project.org/core/dune-common/-/merge_requests/11493.6https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1169Add pipeline in C++20 mode2022-07-27T11:47:40ZTimo Kochtimokoch@math.uio.noAdd pipeline in C++20 modeWe want to support C++20 and use it in one of the upcoming releases. So we should start testing with C++20.We want to support C++20 and use it in one of the upcoming releases. So we should start testing with C++20.3.6Dennis GläserDennis Gläserhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1168New fcstaggered without caching is extremely slow2023-12-13T11:13:10ZTimo Kochtimokoch@math.uio.noNew fcstaggered without caching is extremely slowThere should be some optimization possible. It's really a crazy lot slower than with caching.There should be some optimization possible. It's really a crazy lot slower than with caching.https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1165Separate extended 1.5phase Richards model into a separate model2022-10-05T07:59:44ZTimo Kochtimokoch@math.uio.noSeparate extended 1.5phase Richards model into a separate modelThe option enableDiffusionInGasPhase appears in many places and makes the standard Richards model hard to read.
Per default we also use SwitchablePrimaryVariables which are only needed for the extended Richards model.
I propose to separ...The option enableDiffusionInGasPhase appears in many places and makes the standard Richards model hard to read.
Per default we also use SwitchablePrimaryVariables which are only needed for the extended Richards model.
I propose to separate this into two models. In order to make Richards tests pass that set the phase state we need a deprecation period.3.6Sina AckermannSina Ackermannhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1152Move FECache from gridgeometry to grid variables?2023-12-13T11:12:28ZTimo Kochtimokoch@math.uio.noMove FECache from gridgeometry to grid variables?The FECache in Box is the only type that requires to know about the scalar type. This is because it is supposed to map to variables. So shouldn't it be in the grid variables?
__More details__
The concept of the grid geometry does not h...The FECache in Box is the only type that requires to know about the scalar type. This is because it is supposed to map to variables. So shouldn't it be in the grid variables?
__More details__
The concept of the grid geometry does not have a notion of variables living on the grid and is strictly related to the geometrical description of the domain/grid adapted for a specific discretization scheme. As such it should also only be concerned with the coordinate type and not the scalar variable type.
One element breaking this is the finite element basis in Box which is currently used to interpolate variables. In the concept above it seems ok to use it for coordinate interpolation, but the concept of constructing discrete variable function spaces and bases seems to be what the grid variables are for (mapping variables onto a given grid geometry).https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1147Add pipeline to see if dumux can be used when installed2022-05-04T14:28:20ZTimo Kochtimokoch@math.uio.noAdd pipeline to see if dumux can be used when installedCollect ideas on how to test this. Maybe tests should be a git submodule so they can be checked out easily as part of a testing module as well as dumux itself. The testing module could then use an installed version of dumux in the CI.Collect ideas on how to test this. Maybe tests should be a git submodule so they can be checked out easily as part of a testing module as well as dumux itself. The testing module could then use an installed version of dumux in the CI.https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1141Repartition interface with PTScotch2023-12-13T11:13:09ZTimo Kochtimokoch@math.uio.noRepartition interface with PTScotchThe graph partitioning interface added in !3005 assumes that the initial graph is on rank 0. This is useful to distribute a grid read on rank 0 e.g. with the gmsh reader. However, for dynamic load balancing, we need a repartition interfa...The graph partitioning interface added in !3005 assumes that the initial graph is on rank 0. This is useful to distribute a grid read on rank 0 e.g. with the gmsh reader. However, for dynamic load balancing, we need a repartition interface that works if the graph is distributed. PTScotch provides `dgraph` (distributed graph) for this purpose. Distributed graphs have a more complex interface but include also the current case where the entire graph is on rank 0 (all other ranks then have 0 graph nodes).https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1139Geomechanics documentation is missing the equation overview2022-04-26T11:20:37ZTimo Kochtimokoch@math.uio.noGeomechanics documentation is missing the equation overviewThe current geomechanics models don't state the implemented equations in doxygen.The current geomechanics models don't state the implemented equations in doxygen.3.5Yue Wangyue.wang@iws.uni-stuttgart.deYue Wangyue.wang@iws.uni-stuttgart.dehttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1126Add CO2 benchmark case?2023-12-13T11:13:09ZTimo Kochtimokoch@math.uio.noAdd CO2 benchmark case?Is it possible/feasible to add the benchmark case Holger Class et al , A benchmark study on problems related to CO2 storage in geologic formations, Computational Geosciences 13 (2009), no. 4, 409–434.
into the test suite?Is it possible/feasible to add the benchmark case Holger Class et al , A benchmark study on problems related to CO2 storage in geologic formations, Computational Geosciences 13 (2009), no. 4, 409–434.
into the test suite?https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1112Move porenetwork spatialparams to porenetwork folder2022-02-09T13:23:05ZTimo Kochtimokoch@math.uio.noMove porenetwork spatialparams to porenetwork folderSee !2888
- [x] In the boundary coupled ff_pnm tests, try to reuse the psuedo 3D spatial params used in the psuedo 3d channel freeflow testsSee !2888
- [x] In the boundary coupled ff_pnm tests, try to reuse the psuedo 3D spatial params used in the psuedo 3d channel freeflow tests3.5https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1111Use new spatialparams interface to add spatialparams for freeflow models2022-02-09T13:23:06ZTimo Kochtimokoch@math.uio.noUse new spatialparams interface to add spatialparams for freeflow modelsSee !2888See !28883.5https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1104Multiphase/ Component specific Dispersion tensors2022-03-31T14:16:48ZNed ColtmanMultiphase/ Component specific Dispersion tensors<!--
This form is for feature requests ONLY!
If you're looking for help check out the [readme](/README.md).
-->
**Feature request**
Complex Dispersion tensors
**What does this feature / why does DuMux need it**:
Currently there are di...<!--
This form is for feature requests ONLY!
If you're looking for help check out the [readme](/README.md).
-->
**Feature request**
Complex Dispersion tensors
**What does this feature / why does DuMux need it**:
Currently there are dispersion fluxes implemented for compositional-flow(1pnc) and tracer models. The dispersion tensors are either user defined, or defined using the scheidegger dispersion relation, which is velocity dependent.
Other dispersion tensors exist, and can be component specific or dependent on state and phase conditions. These dipsersion tensors could be added to make the dispersion interface useable in complex cases.3.5Ned ColtmanNed Coltmanhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1097Harmonic mean and other helpers in spatial params2022-03-31T12:29:52ZTimo Kochtimokoch@math.uio.noHarmonic mean and other helpers in spatial paramsThe spatialparams contain an harmonic mean function and a special averaging for tensors. Why is this part of the spatial params interface? It should probably be out-sourced.
The following discussion from !2888 should be addressed:
- [ ...The spatialparams contain an harmonic mean function and a special averaging for tensors. Why is this part of the spatial params interface? It should probably be out-sourced.
The following discussion from !2888 should be addressed:
- [ ] @timok started a [discussion](https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/merge_requests/2888#note_67814): (+3 comments)
> Is this still used? Seems like the wrong place for such a function.3.5Mathis KelmMathis Kelmhttps://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1064Extract module script CMake macros not considered2022-02-23T20:26:33ZTimo Kochtimokoch@math.uio.noExtract module script CMake macros not consideredThe extract module script does not consider cmake macros. So if you extract from a module that defines some cmake macros/functions and then use them, the extracted module will not find these functions because it shouldn't depend on its p...The extract module script does not consider cmake macros. So if you extract from a module that defines some cmake macros/functions and then use them, the extracted module will not find these functions because it shouldn't depend on its parent module anymore.
Maybe we can simply copy the cmake/modules folder (might need some renaming) of the module extracted from.3.5https://git.iws.uni-stuttgart.de/dumux-repositories/dumux/-/issues/1060Simplify localView2021-08-17T17:59:39ZTimo Kochtimokoch@math.uio.noSimplify localViewWe write a lot of code like this
```cpp
auto fvGeometry = localView(gridGeometry);
fvGeometry.bind(element);
```
How about introducing an overload
```cpp
auto localView(g, element) { auto view = g.localView(); view.bind(element); retu...We write a lot of code like this
```cpp
auto fvGeometry = localView(gridGeometry);
fvGeometry.bind(element);
```
How about introducing an overload
```cpp
auto localView(g, element) { auto view = g.localView(); view.bind(element); return view; }
```
so we can write instead (for most cases)
```cpp
auto fvGeometry = localView(gridGeometry, element);
```3.5