Meaning of `A >: Null`?

Question

I attempted to write a function that won't compile if null is passed:

$scala
Welcome to Scala 2.11.8 (Java HotSpot(TM) 64-Bit Server VM, Java 1.8.0_101).
Type in expressions for evaluation. Or try :help.

scala> :t null
Null

scala> def f[A >: Null](x: A):A = x
f: [A >: Null](x: A)A

However, it did not work as I expected:

scala> f( null )
res1: Null = null

Answer 1

As has been pointed out, A >: Null allows for A to be Null itself, but that doesn't even matter here. Even if A were required to be a strict supertype of Null, it would still be possible to pass null to f as null is a valid value of those types (which follows directly from the fact that Null is a subtype of them).

If you don't want to accept nullable types, A needs to be a subtype of AnyVal. Note that this will prevent any values of reference types to be passed - not just null itself.

There is no way to accept reference types without also accepting null.

Answer 2

Scala specification section 3.5.2 (conformance) states (emphasis mine):

The <: relation forms pre-order between types, i.e. it is transitive and reflexive. least upper bounds and greatest lower bounds of a set of types are understood to be relative to that order.

So if we look at the type constructor declaration from a mathematical point of view, we see the :< relation as:

Null <: T <: Any

Since by definition the relation is reflexive, Null <: Null is in the relation, which makes null a valid value to be passed into the method.

As @sepp2k rightly said in the comments and in his answer, the fact that A isn't constrained by an upper bound means that Any is a valid candidate, which means a null value could be passed anyway.