xarray_at_min_of

PlasmaCalcs.tools.xarray_tools.xarray_indexing.xarray_at_min_of(array, of, dim=None, *, promote_dims_if_needed=True)

return array values at locations of minimum of. Roughly: array.isel(of.argmin([dim])).

But, here is nicer, in a few ways:

(1) here doesn’t require array to have all dims in dim.

E.g., if array has dims ‘x’, ‘kmod’, and of has dims ‘kmod’, ‘kang’,

and dim=[‘kmod’, ‘kang’], it will be handled in the intuitive way. Roughly:

amin = of.argmin([‘kmod’, ‘kang’])

result = array.isel(kmod=amin[‘kmod’])

(2) here allows All-NaN slice in of.

all-nan slices in of will be filled with nan values in result.

E.g., if dims=[‘d0’, ‘d1’], of has dims [‘d0’, ‘d1’, ‘nmul’, ‘fluid’],

and of.isnull().all([‘d0’, ‘d1’]) when nmul index == 4 or 5, and fluid index == 2,

then result.isel(nmul=[4,5], fluid=2) will be filled with nan values.

array: xarray.DataArray or xarray.Dataset

array to get result values from.

of: xarray.DataArray

array to find minimum values in.

dim: None, str, or list of str

dim(s) along which to find the minimum values of of.

Takes the “simultaneous” minimum along all these dims, as per argmin(dims)

None –> use of.dims.

(if dim is a list with len==0, return array, unchanged.)

promote_dims_if_needed: bool

whether to promote non-dimension coords to dimensions.

if False, raise DimensionKeyError if any relevant coord is not already a dimension.