- Dockerfile: needs mpirun in the path (0.5)
- Dockerfile: needs to have the proper token stuff for Jupyter (0.5)
- Dockerfile: add the help script etc.
- Separate binder version (branch) (0.6)
- Github actions for testing, deployment / docs
- PyPi wheels
- Conda wheels
- MeshVariable: add print function to output array value
__str__
and__repr__
(0.5) - MeshVariable: add gradient and interpolation / evaluate methods (0.5)
- mesh.add_variable ... cf underworld (0.5)
- eliminate need for mesh in parameter functions (0.6)
- expose DM labels as functions / meshVariables
- dependency tracking (useful for non-linearity detection in solvers) (0.7)
- human-readable naming for fns (meshes / solvers too) (0.7)
- store values at each evaluate statement to reuse (?)
- Sympy integration (for differentiation !) (for 1.x)
- Pixmesh: add an interpolate method.
- Spherical meshing framework (0.8)
- Add scaling module - i.e.
from quagmire import scaling
(0.6) - Integrate scaling into MeshVariable view / getter
- Integrate scaling into meshes
- Save / Load ? Scaling dictionary ?
- Save xmf files for Paraview (0.7)
- Save xmf time series (0.7)
- Save / load project file (
.quag
?) (0.7) - Check if hdf5 is a valid viewer for PETSc (should not crash completely)
- Explicit
- Diffusion (0.6)
- Linear, constant diffusivity + benchmark (0.6)
- Spatially variable diffusity + example (0.6)
- Non-linear diffusivity
- Timestep for non-linear case (meaning of provided step ?)
- Semi-implicit (?)
- Erosion-deposition (0.7)
- Erosion / deposition governing equations + example (0.7)
- Linear, constant erodability (0.7)
- Spatially verying erodability (0.7)
- Integrated erosion-deposition-diffusion time stepping
- Diffusion (0.6)
- Example notebooks with reasonable coverage - all passing
- Unit test scripts (serial)
- Unit test scripts (parallel)
- lavavu examples (0.3)
- lavavu parallel wrappers (cf gLucifer)
- lavavu mesh variable wrappers (cf gLucifer)
- lavavu + shapely for maps (?)