GSOC ideas

Potential mentors

• Nico Schlömer - [email protected]
• Geordie McBain - [email protected]
• Tianyi Li - [email protected]
• Keurfon Luu - [email protected]

Potential projects

• Adding I/O for new formats:

  • glb
  • dxf
  • unv
  • 2dm
  • mfem
  • Silo
  • polyMesh

• Support for multiple meshes per file

• Push completeness and conformity of "power" formats like Exodus and XDMF. These formats have lots of features of potential use to finite element developers that aren't yet supported by meshio. The use-case here isn't so much conversion between formats as serialization and deserialization, and visualization (Exodus & XDMF, e.g., being supported by tools like ParaView & VisIt).

• Completeness and conformity of MSH 4.1. This is an important format because of the position of Gmsh as a mesh-generator.

• Increase test coverage. Right now, we're at 80% with about 1000 lines of code untested. Ansys and XDMF are important offenders.

• Integration of meshio in the scicomp ecosystem. Right now, it's being used by a number of larger projects (e.g., FEniCS), but more packages could profit from meshio integration.

• There are numerous single-purpose readers out there for I/O of a only single file format. Sometimes, they are more capable or faster than meshio. See if we can integrate some of their ideas to make meshio even better.