from_meshio fails on old ex27

I was trying to regenerate a mesh for ex27 as part of the discussion with @bhaveshshrimali on the speed of pypardiso relative to scipy.sparse in #690 so that the number of elements could be varied but the old make_mesh(make_geom(...)) code seems not to work with the new (#680 ?) skfem.io.mesh.from_meshio.

The code:

from skfem import *
from skfem.io.meshio import from_meshio

from itertools import cycle, islice

import numpy as np

from pygmsh import generate_mesh
from pygmsh.built_in import Geometry


def make_geom(length: float, lcar: float) -> Geometry:
    # Barkley et al (2002, figure 3 a - c)
    geom = Geometry()

    points = []
    for point in [
        [0, -1, 0],
        [length, -1, 0],
        [length, 1, 0],
        [-1, 1, 0],
        [-1, 0, 0],
        [0, 0, 0],
    ]:
        points.append(geom.add_point(point, lcar))

    lines = []
    for termini in zip(points, islice(cycle(points), 1, None)):
        lines.append(geom.add_line(*termini))

    for k, label in [
        ([1], "outlet"),
        ([2], "ceiling"),
        ([3], "inlet"),
        ([0, 4, 5], "floor"),
    ]:
        geom.add_physical(list(np.array(lines)[k]), label)

    geom.add_physical(geom.add_plane_surface(geom.add_line_loop(lines)), "domain")

    return geom


if __name__ == "__main__":
    from pathlib import Path

    geom = make_geom(35.0, 0.2)

    with open(Path(__file__).with_suffix(".geo"), "w") as fgeo:
        fgeo.write(geom.get_code())

    mio = generate_mesh(geom, dim=2)
    print(mio)

    mesh = from_meshio(mio)
    print(mesh)

    mesh.save(Path(__file__).with_suffix(".vtk"))

When I run this with pygmsh==6.1.1 (before the big API change in pygmsh 7), I believe that this generates a valid *.geo file, which is accepted by Gmsh (4.8.4 here) and Gmsh can write this as MSH 4.1 file which is understood by meshio, but something goes wrong in skfem.io.meshio.from_meshio:

Traceback (most recent call last):
  File "/home/gdmcbain/src/scikit-fem/docs/examples/ex27_mesh.py", line 55, in <module>
    mesh = from_meshio(mio)
  File "/home/gdmcbain/miniconda3/envs/skfem/lib/python3.9/site-packages/skfem/io/meshio.py", line 93, in from_meshio
    facets = {k: [tuple(f) for f in
  File "/home/gdmcbain/miniconda3/envs/skfem/lib/python3.9/site-packages/skfem/io/meshio.py", line 94, in <dictcomp>
    np.sort(m.cells_dict[bnd_type][v[bnd_type]])]
IndexError: index 381 is out of bounds for axis 0 with size 380

I don’t mean to suggest that we should rely upon pygmsh<7 but I believe the GEO and MSH 4.1 files are valid, as is the meshio.Mesh, but I’ll check.

The intermediate files are generated and interrogated with

python docs/examples/ex27_mesh.py  # say, depending on where the above is saved
gmsh -2 docs/examples/ex27_mesh.geo
meshio info docs/examples/ex27_mesh.msh

[gdmcbain]

Oh, O. K., great. (I wonder what’s going on here, but I’ll sort that out later.) You should be able to change lcar, the second argument to make_geom, to get meshes of different fineness to experiment with in #690.

Are you running on scikit-fem master, i.e. with #680? There were some substantial changes to skfem.io.meshio.from_meshio in that which I don’t know have been published as a release.

[gdmcbain]

I’m not sure why this error didn’t show up earlier. I couldn’t see anything in #680 that would have triggered it. Maybe we were just lucky before and the spurious "gmsh:bounding_entities" were imbibed but happened not to contain invalid indices and so didn’t raise the IndexError and were carried around as worthless but harmless ballast?