As integer will be a "visible" subtype, you can choose not to use
it.
How hard will it be to ensure that all the computations inside a given collection of functions are being done with floating point math? I have a library for generating parametric paths, and I know all of its internal computations ought to be floating point -- both for reasons of speed and precision handling. Will I need to explicitly convert any numeric arguments sent to that library to the float subtype? Something like:
function path(x,y)
return internal_path( tofloat(x), tofloat(y) )
end ?
-Sven
