Narrowing Type Conversion: Why is assignment of int to a byte in a declaration allowed?

Question

This may sound too trivial for an intermediate Java programmer. But during my process of reviewing Java fundamentals, found a question:

Why is narrowing conversion l

Answer 1

Because byte b = 13 ; is assignment of a constant. Its value is known at compile time, so the compiler can/should/will whine if assignment of the constant's value would result in overflow (try byte b = 123456789 ; and see what happens.)

Once you assign it to a variable, you're assigning the value of an expression, which, while it may well be invariant, the compiler doesn't know that. That expression might result in overflow and so the compiler whines.