_Unsigned _Bit discrepancy #50
Description
When doing calculations with _Unsigned _Bits, if the number goes negative, QB64 acts inconsistent from how it acts with _Unsigned _Integers.
Whether this is intentional or not, I don't know. However, it appears to me like a bug.
Run the program at the bottom to see what I mean.
Running an _Unsigned _Bit and and _Unsigned _Integer, set to the same value, through the same calculation, yield vastly different results!
Normally QB64 does a sound job of returning a result in the form of the least common denominator variable type. For example, in the calculation, "3 / 2" a "1.5" would be returned. Or another example, "2 - 3" would return -1. Exceptions to this are when the calculation is assigned to a variable. Then the assigned variable's type is used. This is expected behavior.
Now consider the _Unsigned _bit:
In equations resulting in a decimal, nothing unusual happens.
But when things turn negative...
The rules of _Unsigned _Integer variable type seems to be used. That doesn't seem quite right.
For example, "0 - a" where a is dimensioned as _Unsigned _Bit, 4294967295 is returned. Which is -1 for _Unsigned _Integers
Shouldn't, at the very least, _Unsigned _Bit rules of calculation be used, so it would return the top value given the amount of bits dimmensioned?
However, try the same equation with an _Unsigned _Integer, and you get -1!
I believe the way _Unsigned _Integers work is the way _Unsigned _Bits should as well.
Dim UnsignedBIT As _Unsigned _Bit
Dim UnsignedINT As _Unsigned Integer
Dim SignedINT As Integer
UnsignedBIT = 1
UnsignedINT = 1
Print "UnsignedBIT = "; UnsignedBIT;
Print "UnsignedINT = "; UnsignedINT
Print ""
Print "Unsigned bit does not act like unsigned Integer."
Print "-UnsignedBIT = "; -UnsignedBIT
Print "-UnsignedINT = "; -UnsignedINT
Print "0 - UnsignedBIT = "; 0 - UnsignedBIT
Print "0 - UnsignedINT = "; 0 - UnsignedINT
Print ""
Print "Saving to a signed bit works as expected."
SignedINT = -UnsignedBIT
Print "SignedINT = -UnsignedBIT"; SignedINT;
End