Randall J. LeVeque

I added another suggestion to https://github.com/clawpack/amrclaw/issues/262 that would be useful in particular for GeoClaw, namely introducing `amrdata.amr_levels_max_outside_flagregions`. That reminded me of another GeoClaw parameter that often causes confusion and/or unexpected...

@donnaaboise - That is exactly what I was suggesting in proposing we allow the user to specify a flagregion with something like: flagregion.spatial_region_type = 3 # topofile extent flagregion.spatial_region_file =...

@mandli: You mean the new `flagregions.data`? It was to accommodate the different set of input data needed for ruled rectangles and other options that can now be set related to...

Yes, it would be nice to have other plotting tools too, particularly for amrclaw simulations that have nothing to do with geosciences!

It's in the examples version.

And also in `tests/multilayer`

The PR on riemann still doesn't explain why things are set differently in these two examples, both of which have `coordinate_system == 1`. I guess they should both be ```...

We have a test in `tests/netcdf_test`, what we don't have is NetCDF installed on Travis.

Note that `$CLAW/amrclaw/src/2d/flagregions2.f90` appears to flag based on regions and so does `$CLAW/geoclaw/src/2d/shallow/flag2refine.f90`. @mjberger: Are both being used? Also some geoclaw/example Makefiles include both `flagregions2.f90` and `flagregions.f` and should be...

You are right that in general the ghost cell values are never used in plotting or other postprocessing, and in fact they are stripped off when using the standard pyclaw...