jcgraciosa

Pytest for vector advection with semi-Lagrangian

3

I've created a small test for verifying the advection of a vector field using the semi-Lagrangian method. This test is similar in structure to test_1100_AdvDiffCartesian.py.

Clarification re. duplicate additions of the boundaries

1

Both essential and natural BCs are added (using `PetscDSAddBoundary_UW`) in the `setup_discretisation` method defined in petsc_generic_snes_solvers.pyx. In turn, the `setup_discretisation` method is called in `build` (or `rebuild_after_mesh_update`) function. This `build`...

Change in return type of getField() function in PETSc version 3.22.1

5

Branch tested: development PETSc version: 3.22.1 Python version: 3.12.7 Issue description: Accessing the swarm fields using ```getField()``` no longer returns the field of interest as an array. Instead, it returns...

[BUG] - Order of MeshVariable declaration becomes important when adding B.C. using add_condition()

3

Branch tested: development PETSc version: 3.22.0 Issue description: When adding boundary conditions with the ```add_condition``` function, the user must declare the MeshVariables used in the solver first in order to...

What is the defined output of <mesh>.points_in_domain() when input is empty?

2

What should we define as the output of .points_in_domain() when the input is empty (i.e. [])? Currently, it returns False. Do we stick with this behaviour? This is relevant when...

What is the defined way of running <basic swarm>.add_particles_with_global_coordinates() in parallel?

1

Suppose we add three particles using ```add_particles_with_global_coordinates()```: ```python xy_st = np.array([[0.5, 0.33], [1.5, 0.33], [1.5, 0.50]]) basic_swarm.add_particles_with_coordinates(xy_st, migrate = True) ``` In serial, this is not a problem. In parallel,...