Skip to content

Latest commit

 

History

History
46 lines (38 loc) · 2.95 KB

README.md

File metadata and controls

46 lines (38 loc) · 2.95 KB
Raw

Examples of CoMMA python module

Dependencies

In most of the scripts here, we use the string format {var=}, hence python3.8 or later is needed.

Two main libraries are used, both are available on GitHub under the MIT license:

  • meshio for reading and exporting meshes in various formats
  • dualGPy for building the dual graph of a mesh

Description

All the examples in the directory have, roughly speaking, the same structure:

  1. Create / Read a mesh
  2. Prepare the mesh with dualGPy
  3. Agglomerate with CoMMA
  4. Export the results to .vtu thanks to meshio

A brief description of the scripts in this directory:

  • ex_default_dualGPy.py: Build a Cartesian 2D mesh with dualGPy default constructor. Simple switches enable the user to choose anisotropy, dimension and other parameters for CoMMA. Useful for familiarizing with:
    • Prepare a mesh with dualGPy so that it can be handled by CoMMA
    • Arguments and outputs of CoMMA call
    • Preparing the output (subdivision in classes) for meshio
  • ex_read_mesh.py: Read an existing mesh via meshio and agglomerate it. Simple switches enable the user to choose anisotropy, dimension and other parameters for CoMMA. Useful for familiarizing with:
    • Reading a mesh with meshio and preparing it with dualGPy
    • Preparing the output (subdivision in classes and/or in cell types) for meshio
  • ex_aniso_lines.py: Show how to draw anisotropic lines obtained from CoMMA. Useful for familiarizing with:
    • CoMMA's definition of anisotropic lines
    • Drawing anisotropic lines with dualGPy
  • ex_several_agglo_levels.py: Show how to agglomerate several times the same mesh using CoMMA. Useful for familiarizing with:
    • Running several agglomeration steps on the same mesh, each time preparing the result of the previous step for the following one
    • Outputting more than one field via meshio
  • ex_rae.py: Agglomeration of a 2D RAE2822 mesh. We use here different features:
    • Perfectly selection of the BL by looking at the cell-type (only squares)
    • Computation of the (reciprocal) of the wall-distance to be used as weights to give higher priority to cells near the wall
    • Several agglomeration levels
    • Modification of the anisotropic lines after the first level

Deprecated

Some old deprecated examples:

  • test_from_comma_source.py: Build a hand-picked 2D mesh, agglomerate it with CoMMA and export the result
  • test_dualGPy.py: Build a hand-picked 2D mesh of quads and triangles with dualGPy, agglomerate it with CoMMA and export the result
  • test_square.py: Build a Cartesian 2D mesh with dualGPy default constructor, agglomerate it with CoMMA and export the result

Formatting

The python files are formatted with Black and isort. A configuration file is available, see pyproject.toml.