JuliaArrays / AxisArrays.jl

Performant arrays where each dimension can have a named axis with values
http://JuliaArrays.github.io/AxisArrays.jl/latest/
Other
200 stars 41 forks source link

In join, allow arbitrary/missing fillvalue type #159

Open gajomi opened 5 years ago

gajomi commented 5 years ago

Currently in calls to join the fillvalue must match the eltype of the arrays that are being joined. If an arbitrary fillvalue type were allowed and some kind of eltype promotion logic added one would have the option of using the much beloved missing as the fillvalue, which is maybe nice for workflows being ported from the various dataframes and/or Python's xarray.DaraArray.