Hi,
I have a code very similar to the one I posted here.
At some point, I compute a heat flux as:
# Heat flux.
Q = functionspace(mesh, ("DG", 0, (mesh.geometry.dim,)))
q = Function(Q)
flux_calculator = Expression(-κ * grad(temp), Q.element.interpolation_points())
q.interpolate(flux_calculator)
Where temp comes from:
# Define ME function space
deg = 1
el = FiniteElement("CG", mesh.ufl_cell(), deg)
mel = MixedElement([el, el])
ME = FunctionSpace(mesh, mel)
u, v = TestFunction(ME)
TempVolt = Function(ME)
temp, volt = split(TempVolt)
As far as I understand, the above code should work. However, I get an error when I save the heat flux field with:
with VTXWriter(MPI.COMM_WORLD, "results/heat_flux.bp", [out2], engine="BP4") as vtx:
vtx.write(0.0)
The traceback is:
Traceback (most recent call last):
File "/usr/local/dolfinx-real/lib/python3.10/dist-packages/dolfinx/io/utils.py", line 91, in __init__
comm, filename, output._cpp_object, engine) # type: ignore[union-attr]
AttributeError: 'list' object has no attribute '_cpp_object'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "/root/shared/TE_simulation/BN_example.py", line 261, in <module>
with VTXWriter(MPI.COMM_WORLD, "results/heat_flux.bp", [out2], engine="BP4") as vtx:
File "/usr/local/dolfinx-real/lib/python3.10/dist-packages/dolfinx/io/utils.py", line 94, in __init__
self._cpp_object = _vtxwriter(comm, filename, _extract_cpp_functions(
RuntimeError: VTK does not support cell-wise fields. See https://gitlab.kitware.com/vtk/vtk/-/issues/18458.
One way to bypass this problem is to pick a higher degree for the heat flux (which I don’t think make sense). It doesn’t make sense because the temp field is CG 1st degree, so linear. The heat flux involves a spatial derivative of it, so it should be cell-wise constant, and DG (because I picked it this way). Therefore, picking a deg == 1 is not correct. But this fixes this error… What is going on?
I also got strange warnings when I played around with the degrees of both temp and the heat flux, such as:
libffcx_forms_6d805b63b192afc755cfe28f475c8e3fdb08abc5.c: In function ‘tabulate_tensor_integral_888714572bbf59b20e226d6f0e3ddb633201b4cc’:
libffcx_forms_6d805b63b192afc755cfe28f475c8e3fdb08abc5.c:855:21: warning: unused variable ‘weights_083’ [-Wunused-variable]
855 | static const double weights_083[1] = {0.5};
| ^~~~~~~~~~~
At top level:
libffcx_forms_6d805b63b192afc755cfe28f475c8e3fdb08abc5.c:855:21: warning: ‘weights_083’ defined but not used [-Wunused-const-variable=]
libffcx_forms_1bccbd7aa7954a36b415fefd115e419ef5f34d0f.c: In function ‘tabulate_tensor_integral_d8173cb13faee8cb1f81d5db877573e52f4a8903’:
libffcx_forms_1bccbd7aa7954a36b415fefd115e419ef5f34d0f.c:855:21: warning: unused variable ‘weights_083’ [-Wunused-variable]
although I don’t get those with the code above. I do not remember which combination of degrees yielded these warnings.