casting

[prakash s]

by default every integer literal is an int
so how does this works witjout typecasting
byte b=6;

[Val Dra]

it looks at a value of 6 and sees that it's in a range of byte so no casting is necessary. But if you do this
byte b = 6;
b = b + 1; // Error because binary operation causes an error you try to add 1 which is an int and the result is not in the range of byte so you must cast
operations like this b++; will not caouse an error because implicit cast is being performed.

[puneet pruthi]

Hi!!
constants which r in range wont throw any error.like byte b=10;though 10 is in range ,still the compiler knows that it is a constant whose value is not going to change and it is within limits so it is okie
another thing which will work is
final int j=30;
byte b=j;
this will also work fine.

[Lafayette Hubert]

So why doesn't this work with floats/doubles, etc?
e.g., why can't I do:
float f = 3.14; //clearly within range & precision of f.

Originally posted by puneet pruthi:
Hi!!
constants which r in range wont throw any error.like byte b=10;though 10 is in range ,still the compiler knows that it is a constant whose value is not going to change and it is within limits so it is okie
another thing which will work is
final int j=30;
byte b=j;
this will also work fine.

[Ishaan Mohan]

Rules apply on char,short and byte. If the integer literal is in range of the given primitive, compiler allow narrowing cast.
float f=3.14; here 3.14 is double and you cannot assign it without explicit cast. If expression is float f=30; compiles fine.