Fix bug with raw fields when there is no ghost cells

[NeilZaim]

I was trying to read raw fields from a plotfiles coming from a WarpX simulation and I encountered a bug occuring when there is no ghost cells. The bug occurs when doing something like:

import yt
ds = yt.load("diags/diag100200")
ad = ds.all_data()
Ex_raw = ad['raw','Ex_aux']

And the error message is:

Traceback (most recent call last):
  File "./analysis_reduced_diags.py", line 58, in <module>
    Ex_raw = ad['raw','Ex_aux'].to_ndarray()[:,0]
  File ".../yt/data_objects/data_containers.py", line 256, in __getitem__
    self.get_data(f)
  File ".../yt/data_objects/data_containers.py", line 1571, in get_data
    fluids, self, self._current_chunk)
  File ".../yt/geometry/geometry_handler.py", line 244, in _read_fluid_fields
    chunk_size)
  File ".../yt/frontends/boxlib/io.py", line 66, in _read_fluid_selection
    nd = g.select(selector, ds, rv[field], ind)
  File ".../yt/data_objects/grid_patch.py", line 421, in select
    dest[offset:offset+count, i] = source[sl][np.squeeze(mask)]
IndexError: boolean index did not match indexed array along dimension 0; dimension is 0 but corresponding boolean dimension is 32

The bug happens with yt 3.6 but the same code was running well with version 3.5.1.

After backtracing the error, it seems to be coming from using slice(nghost[dim],-nghost[dim]) to trim the ghost cells in the function _read_raw_fields, which fails when nghost[dim]=0. I propose a simple fix in this PR which worked for me.

[welcome]

Hi! Welcome, and thanks for opening this pull request. We have some guidelines for new pull requests, and soon you'll hear back about the results of our tests and continuous integration checks. Thank you for your contribution!

[atmyers]

This looks good to me - thanks for the catch and the fix!

[welcome]

Hooray! Congratulations on your first merged pull request! We hope we keep seeing you around! 🎆