hypothesis types strategies should accept constraints

[averagehat]

e.g.

type_to_strat(SettingsObj, { 'x' : {'min_size' : 3 } })

[averagehat]

Alternative: use .filter and the name of the type. The filter gets applied after the construction of that type. i.e.

type_to_strat(GianRobot, { Point3D : lambda p: p.x >= 0 and p.y >= 0 and p.z >= 0 } )

results in

Point3D(x,y,z).filter(lambda p: . . . . 

This would be trivial if we could define some public constructor for Point3D which returned Either or Option, but I don't believe it's possible to make the type of __new__ for class T something other than T