Mesh cannot run due to TypeError

Errors
,
1

Dear all,
my dolfinx.mesh cannot work even for the simplest unit interval mesh generation. Here is the code:

 from mpi4py import MPI
 import numpy as np
 import ufl
 from dolfinx import mesh
 comm = MPI.COMM_WORLD
 nx = 20
 msh = mesh.create_unit_interval(comm,nx)

and here is the error message:

Exception has occurred: TypeError
create_interval_float64(): incompatible function arguments. The following argument types are supported:
1. create_interval_float64(comm: MPICommWrapper, n: int, p: collections.abc.Sequence[float], ghost_mode: dolfinx.cpp.mesh.GhostMode, partitioner: collections.abc.Callable[[MPICommWrapper, int, collections.abc.Sequence[dolfinx.cpp.mesh.CellType], collections.abc.Sequence[numpy.ndarray[dtype=int64, writable=False]]], dolfinx.cpp.graph.AdjacencyList_int32] | None) → dolfinx.cpp.mesh.Mesh_float64

Invoked with types: Intracomm, int, list, dolfinx.cpp.mesh.GhostMode, NoneType
File “/home/yjin6/Work/FEnicsX demo/import ufl.py”, line 8, in
msh = mesh.create_unit_interval(comm,nx)
TypeError: create_interval_float64(): incompatible function arguments. The following argument types are supported:
1. create_interval_float64(comm: MPICommWrapper, n: int, p: collections.abc.Sequence[float], ghost_mode: dolfinx.cpp.mesh.GhostMode, partitioner: collections.abc.Callable[[MPICommWrapper, int, collections.abc.Sequence[dolfinx.cpp.mesh.CellType], collections.abc.Sequence[numpy.ndarray[dtype=int64, writable=False]]], dolfinx.cpp.graph.AdjacencyList_int32] | None) → dolfinx.cpp.mesh.Mesh_float64

Invoked with types: Intracomm, int, list, dolfinx.cpp.mesh.GhostMode, NoneType

I have tested my py4mpi that worked well. I am very confused which argument type is not accepted for the mesh generation.

2

Your code is correct, at least for v0.10. Could you provide the necessary details to understand your setup? Which version are you running, how did you install things, etc.

3

Hello Stein,
Thank you very much for your reply. Sorry about my silly question. I downloaded the main branch which is under development. When I installed the release version, I did not meet mesh issue anymore. I will close this question.

4

The code should still run on the main branch, but I guess you are mixing conda and installing from source?

5

Hello Dokken,

Thanks for your reply. I found the problem comes from the function handle difference due to multiple compiler (gcc/intel icx/gfort/intel ifx). That makes the _cpp class cannot figure out the argument transferring. I am still configuring the whole python interface including mpi4py\petsc4py\slepc4py and fenics .etc.