Is this how to rename boundaries and subdomains generated in GMSH

Question

I've been reading through the documentation and I haven't been able to successfully solve the test problem I've been trying to implement using a mesh I generated in GMSH. Rather than renaming the surfaces and volumes in the .geo file, I want to do this post-mesh generation and conversion using dolfin-convert. Is it ok to rename these subdomains and boundaries and define the surface/volume differentials as I have below? I'm trying to determine if this is the issue in my code or if I must look elsewhere for additional problems. If it helps, my domain has two Dirichlet BC's on domain 34 (30 and 31 from the .msh file) and an interior interface between subdomains 33 and 34 with a flux continuity Neumann BC on it. Subdomain 33 is entirely encompassed by subdomain 34.

mesh = Mesh(ofilename)

subdomains = MeshFunction("size_t", mesh, "%s_physical_region.xml"%ofilename.split('.')[0])

boundaries = MeshFunction("size_t", mesh, "%s_facet_region.xml"%ofilename.split('.')[0])

fix_subdomains = CellFunction("size_t", mesh)

fix_subdomains.set_all(0)

fix_subdomains.array()[subdomains.array()==34] = 1

fix_subdomains.array()[subdomains.array()==33] = 2

fix_boundaries = FacetFunction("size_t", mesh)

fix_boundaries.set_all(0)

fix_boundaries.array()[boundaries.array()==30] = 1

fix_boundaries.array()[boundaries.array()==31] = 2

fix_boundaries.array()[boundaries.array()==32] = 3

fix_interface = FacetFunction("size_t", mesh)

fix_interface.set_all(0)

fix_interface.array()[boundaries.array()==32] = 1

dx = Measure("dx", subdomain_data=fix_subdomains)

ds = Measure("ds", subdomain_data=fix_boundaries)

dS = Measure("dS", subdomain_data=fix_interface)

Answer 1

As an update, I found another issue with the way I was identifying the physical volumes in GMSH that was the problem. It appears that this scheme does work to rename numbered boundaries and subdomains.

Comments

I have had trouble identifying physical surfaces and volumes when I extrude a 2-D geometry, I just can't recover the physical groups from the .geo file. If I build the geometry from the GUI I can see the groups labeled properly, but using the mesh generated from there has never transfered to DOLFIN properly. Would you mind sharing your GMSH problem and resolution?

---

Maybe it would help to see what I did in the code and the .geo and .msh files and how those turned out. I'm new to FEniCS, but maybe it'll help. If it's relevant, I'm using FEniCS in docker currently because that's the only way I can get dolfin-convert to function correctly.

Here's a link to the necessary code and files: https://github.com/sweeneyd/FEniCS-Examples/blob/master/goodCube/goodCube.py