#426 Merged at e58dbfd
Repository
clean-copy
Branch
week-of-code
Repository
enzo-dev
Branch
week-of-code
Author
  1. John Wise
Reviewers
Description

These changes enable all or a subset of the split particles to be marked as must-refine particles that can then take advantage of its machinery, either restricting AMR to their Lagrangian volume and/or refining to some AMR level.

  • Adding an option to make split particles into must-refine ones
  • Adding additional particle splitting functionality from John Regan's enzo-3.0 fork.
  • Correcting float macro for particle splitter region center
  • Additional MRP check to not include DM particles that were created from old star particles and merging routines.
  • Adding capability to read particle IDs to mark as must-refine during particle splitting.
  • Don't rebuild hierarchy before particle splitting. Will lose information if using must-refine particles.
  • Additional restrictions on split must-refine particles
  • More debugging of split must-refine particles
  • Adding must-refine particle splitting parameters to docs

Comments (9)

  1. John Regan

    @jwise77 This looks good. I think we might need some examples of how to set up the Langranian region for when using ParticleSplitterMustRefineIDFile otherwise I can people getting this wrong. Did you use some python scripts to extract the particle ids that need to be split? Maybe we need a section in the docs about this?

    1. John Wise author

      Thanks. I used my set of MUSIC / yt tools (enzo-mrp-music) to generate the particle list. I think it’s a good idea to provide a small yt sample script (without using my tools) to generate such a file. I’ll update the docs with it.

    2. John Wise author

      Hi @john_regan I’ve updated the docs to contain a short sample script to create the file.

  2. Greg Bryan

    This looks good to me! Just one comment (in addition to Britton’s points) I think the documentation should make clear that there are two ways to specify the splitting region and that they both shouldn’t be used (if I understand correctly).

  3. John Wise author

    I’ve addressed all of your feedback. I think it’s good to go now. Sorry for the huge delay!