AbstractFloat

[ChrisRackauckas]

Why the restriction to AbstractFloat?

https://github.com/shivin9/PDEOperators.jl/blob/master/src/fornberg.jl#L4

Usually the coefficients are integers and rationals. Restricting them to float seems a little odd. This can just be determined from the grid step.

[ChrisRackauckas]

You'd have to make sure you "free-range" the other Ts too.

[shivin9]

ok I'll check on that. @dextorious might have had some reason for defining it that way.

[dextorious]

The only reason I had was that I was only working with floating point data and wanted an easy guarantee of type stability - the code wasn't originally written to be part of a general library. In other words, I was lazy. :)

@ChrisRackauckas is right in saying that the type can just be determined from the grid step, AbstractFloat is unnecessarily restrictive in this case.

[shivin9]

Fixed in f1985bc