eqiora.meshing

← Python API index

Plan and generate meshes from geometry.

View source

Example: mesh the rectangle

Continue with geometry from the geometry example:

python

from eqiora import meshing

mesh_plan = meshing.resolve(geometry, meshing.CartesianMesher(cells=(8, 4)))
mesh = meshing.generate(mesh_plan)
print(mesh.dimension, mesh.cell_count, mesh.vertex_count)
print(mesh.coordinates.shape)

The rectangle becomes 32 cells with 45 vertices in two dimensions. coordinates contains the vertex positions. Use this same mesh when resolving a spatial model bound to this geometry; see modeling.

API

AffineTriangleMesher

Mesh a rectangle with a structured grid of affine triangles.

Each grid cell is split along its lower-left to upper-right diagonal. You choose only the (positive) number of subdivisions in each direction.

python

@final
class AffineTriangleMesher:
    def __new__(cls, *, cells: tuple[int, int]) -> Self: ...
    @property
    def cells(self) -> tuple[int, int]: ...
    @property
    def diagonal(self) -> str: ...
    def __eq__(self, other: object, /) -> bool: ...
    def __repr__(self) -> str: ...

CartesianMesher

Mesh a Cartesian domain with a structured grid.

python

@final
class CartesianMesher:
    def __new__(cls, *, cells: tuple[int] | tuple[int, int] | tuple[int, int, int]) -> Self: ...
    @property
    def cells(self) -> tuple[int] | tuple[int, int] | tuple[int, int, int]: ...
    def __eq__(self, other: object, /) -> bool: ...
    def __repr__(self) -> str: ...

GmshMesher

Generate the mesh with Gmsh (requires the gmsh extra).

python

@final
class GmshMesher:
    def __new__(cls, *, maximum_boundary_error: float=..., maximum_target_size: float | None=..., minimum_mean_ratio: float=..., maximum_boundary_facets: int=...) -> Self: ...
    @property
    def maximum_boundary_error(self) -> float: ...
    @property
    def maximum_target_size(self) -> float | None: ...
    @property
    def minimum_mean_ratio(self) -> float: ...
    @property
    def maximum_boundary_facets(self) -> int: ...
    def __eq__(self, other: object, /) -> bool: ...
    def __repr__(self) -> str: ...

Mesh

Immutable mesh, linked to the geometry and plan that produced it.

python

@final
class Mesh:
    @staticmethod
    def coordinate_factors(model: Model, domain: DomainRef, cells_per_factor: list[int]) -> Mesh: ...
    @staticmethod
    def from_bytes(data: bytes) -> Mesh: ...
    def to_bytes(self) -> bytes: ...
    def refine(self, marked_cells: Iterable[int], /, *, maximum_cells: int) -> Mesh: ...
    @property
    def source_digest(self) -> str: ...
    @property
    def realized_geometry_digest(self) -> str | None: ...
    @property
    def digest(self) -> str: ...
    @property
    def correspondence_digest(self) -> str | None: ...
    @property
    def production_lineage_bytes(self) -> bytes: ...
    @property
    def production_lineage_digest(self) -> str: ...
    @property
    def canonical_bytes(self) -> bytes: ...
    @property
    def dimension(self) -> int: ...
    @property
    def vertex_count(self) -> int: ...
    @property
    def cell_count(self) -> int: ...
    @property
    def coordinates(self) -> npt.NDArray[np.float64]: ...
    @property
    def cells(self) -> npt.NDArray[np.uint32]: ...
    def entity_vertices(self, entity: tuple[int, int]) -> tuple[int, ...]: ...
    @property
    def minimum_mean_ratio(self) -> float: ...
    @property
    def selection_names(self) -> tuple[str, ...]: ...
    def selection_entity_count(self, name: str | GeometrySelection) -> int: ...
    def selection_entities(self, name: str | GeometrySelection) -> tuple[tuple[int, int], ...]: ...
    def __repr__(self) -> str: ...

MeshPlan

Fully resolved meshing plan for one geometry.

python

@final
class MeshPlan:
    @property
    def source_digest(self) -> str: ...
    @property
    def provider(self) -> AffineTriangleMesher | CartesianMesher | GmshMesher: ...
    def __repr__(self) -> str: ...

generate

Run a resolved meshing plan and return the mesh.

python

def generate(plan: MeshPlan, /) -> Mesh: ...

resolve

Resolve a meshing request into a plan for the given geometry.

python

def resolve(geometry: Geometry, provider: AffineTriangleMesher | CartesianMesher | GmshMesher, /) -> MeshPlan: ...