Resolve Java method with omitted varargs parameter

Question

I have the following classes definition:

class A {
String foo(Object par) {
     return par.toString();
  }
}
class B extends A {
 static String foo(String par, Object ... pars) {
   return par.toString();
 }
}

If I do a call like B.foo("hello"); , then I get a compilation error

error: non-static method foo(Object) cannot be referenced from a static context

Of course B.foo("hello", (Object[])null); works just right. However if I rename the virtual method foo, to foo1 for example, then the compilation error gets gone. It means that first variant of a static method call is still valid, however a compiler tries first to match virtual signature. Can someone provide a right explanation the behavior from the Java language specification document?

Answer 1

The method resolution algorithm is defined in the JLS #15.2.2 . In substance, methods without varargs have priority over methods that have varargs (emphasis mine):

The first phase (§15.12.2.2) performs overload resolution without permitting boxing or unboxing conversion, or the use of variable arity method invocation . If no applicable method is found during this phase then processing continues to the second phase.

As explained in the JLS, it is done that way " to ensure compatibility with versions of the Java programming language prior to Java SE 5.0 ".

Answer 2

A.foo is NOT static while B.foo is.

Calling B.foo("Hello") is an error because inherited A.foo method is NOT static. The calls that you say are working probably shows you a warning?