Question about implicit promotion

[k reeta]

Hi,

I am having troubling trying to figure out implicit promotions - since int and float are 32-bit, I assumed it'd be okay to assign one to another. It works for assigning an int to a float e.g.

int y=5;
float f=y;

But not the other way around, ie when I try to assign a float to an int I get a compiler error -

float ff=5.0f;
int x=ff;

Why should this be so since both, float and int are 32-bit??

Thanks.

[Henry Wong]

k reeta wrote:
Why should this be so since both, float and int are 32-bit??

It's not based on the number of bits. It's based on the range. Float has a much larger range than an int, so implicit casting from int to float is okay, but not in the other direction.

Henry

[Venkata Raaman]

k reeta wrote:Hi,

I am having troubling trying to figure out implicit promotions - since int and float are 32-bit, I assumed it'd be okay to assign one to another. It works for assigning an int to a float e.g.

int y=5;
float f=y;

But not the other way around, means when I try to assign a float to an int I get a compiler error -

float ff=5.0f;
int x=ff;

Why should this be so since both, float and int are 32-bit??

Thanks.

Here is the order in ascending:
BYTE <SHORT <INT <LONG<FLOAT<DOUBLE.....

Smaller value casted implicitly....(check your first example: float is greater so assigning int value to float -OK)
But not the other way around.
If you want to assign float value to int do explicit cast.
float ff=5.0f;
int x = (int)ff; int is lower in the order and you are assigning higher element to lower one

[k reeta]

Thank You for the responses, I think I've got it now!

[sakthi moorthy]

Type wise
BYTE <SHORT <INT<LONG<FLOAT<DOUBLE.....
is correct

size wise
BYTE<SHORT<INT<FLOAT<LONG<DOUBLE.....
is correct


Conversion of an int or a long value to float, or of a long value to double, may result in loss of precision-that is, the result may lose some of the least significant bits of the value.
Refer JLS 5.1.2