short a = 32767;
short c = 32767;// cc must be less than 32767
In this case, following statement make syntax error because "a + a " become int type.
c = a + a;
However, following statement does not make syntax error
c += a;
I guess the reason is "+= overloading method" has short type input parameters and return value.
I want to make sure what I knew. Does anyone know where I can get clear reference?
For a "plus" expression (+) with two operands of primitve numeric type, the result is:
1. of a primitive numeric type
2. at least an int because of normal promotions
3. of a type that is as wide as the wider of the two operands.
4. of a value that is calculated by promoting the operans to the result type and performing the addition using those types.
There are a couple of things going on here.
First, the binary addition operator promotes its operands to either int, long, float, or double, depending on the highest-precision operand. The JVM does not provide arithmetic operations for bytes or shorts; they're promoted to ints.
Ref: http://java.sun.com/docs/books/jls/second_edition/html/expressions.doc.html#13510
Second, the compound assignment statement is actually translated like this:
E1 = (T)((E1) op (E2))
In other words, the result of the operation has an implied cast to the type of the result.
Ref: http://java.sun.com/docs/books/jls/second_edition/html/expressions.doc.html#5304
