Hex representation of negative integers in Vala

[Johan_Mattsson]

Hi. I am not sure if this is a bug or a feature but I thought it would be a good idea ask here because Vala handles the negative sign in a way that seems to be different from C.

This statement in C works for me:

   int32_t x = 0xffffcccc;
   assert (x == 0xffffcccc);

but the same thing in Vala causes me some trouble:

	// fiddle with bits, the details are not important but this function will return 0xffffcccc (in this example).
	int32 to_fixed (double d) { 
		int32 val = (int32) Math.floor (d);
		int32 mant = (int32) Math.floor (0x10000 * (d - val));
		val = (val << 16) | mant;
		return val;
	}
	
	void test () {
		int32 f;

		f = 0xffffcccc; // does not compile because "Assignment: Cannot convert from `int64' to `int32'"
		assert (f == 0xffffcccc);
				
		f = to_fixed (-0.2); // this will set f to 0xffffcccc (it does compile)
		
		// this will print "0xffffcccc == 0xffffcccc ?"
		stdout.printf ("0xffffcccc == 0x%x ?\n", f); 
		
		// will fail, f is not 0xffffcccc 
		assert (f == 0xffffcccc); 	
	}

Is there a way around this?

[lwildberg (Lorenz Wildberg)]

I guess the expression 0xffffcccc returns a 64 bit integer. This is why you cannot assign it to a int32 variable. And I think the assertion fails, because f will be casted to int64. Can you maybe cast 0xffffcccc explicitly to int32?

[Johan_Mattsson]

Thank you for taking the time to respond this. Yes, it looks like explicitly casting to int32 works, maybe a warning for the assert case is enough to make sure that this doesn’t happen to other developers.

assert (f == 0xffffcccc);

Cheers