Wrap up

• Matt's notes: Callback from BoxLib to Python for tagging now works!
• Donna: VisClaw sort of works with ForestClaw output. Next: mapped grids embedded in 3D space.
• Randy: working on several things:
  • Community page for docs
  • Cake on the home page!
  • specification of non-rectangular refinement regions (in progress)
  • Leaflets for maps in notebooks (with Chris, in progress)
• Hugo: using GeoClaw for research
• Jed:
  • A working example of IMEX timestepping where PETSc handles the time stepping (link?)
  • Getting this into the library code (inside evolve_to_time; in progress)
  • Next step: an interface where PyClaw controls time stepping and calls PETSc SNES
  • Preparing a CEP for Riemann solver registration
• Yiannis:
  • step-size control logic in PyClaw; still debugging
  • improvement to deepcopy for state objects
• Chris:
  • Plan to integrate leaflets and refinement regions into Chile notebook as an example
• Aron: successfully built stacks on most participant laptops
• David K:
  • Successfully plotted AMRClaw output with yt. Code is here. A VisClaw implementation is future work.
  • Wrote a very simple IMEX solver for viscous Burgers' equation in this branch
  • I got BoxClaw running on my own machine, thanks to HashDist!
  • A bunch of minor improvements to the docs, visclaw, and pyclaw
• Kyle:
  • Wrote an IO plugin to allow at least preliminary plotting for ForestClaw with VisClaw
  • Wrote a backwards compatible point-wise Riemann solver interface
  • Started work on a refactor of the GeoClaw Riemann solver with the intention of using it as a testing ground for the new point-wise solvers including reuse of a 1D point-wise solver for higher dimensions and interoperability with GeoClaw and PyClaw.
  • Simple test example of a BoxClaw flagging criteria
  • Miscellaneous bug fixes