PureBasic Forums - English

when having a value like $096C also 2412, how can i split this to have two variable handling the first and second byte? Also as result, something like:
result1 and result2 needed as .w!!! Is there any way to do this with LSL, LSR (bitshifts) or any math calculations but without any string operations? I know its possible with string operations but still search a way to realise it on another way (math).

Ralf wrote:when having a value like $096C also 2412, how can i split this to have two variable handling the first and second byte? Also as result, something like:

Code: Select all

input.w = 2412
result1.w  = 24
result2.w  = 12

result1 and result2 needed as .w!!! Is there any way to do this with LSL, LSR (bitshifts) or any math calculations but without any string operations? I know its possible with string operations but still search a way to realise it on another way (math).

result1=input>>8
result2=input&$FF

> result1=input>>8

You have to binary 'AND' this also, as PB uses Aritmethic shift, which means that if it's a negative number '1':s will be shifted in from the left...
So this would be the correct way to get the upper byte value:

Having everytime 9 and 108 as result instead 24 and 12 (input.w = 2412) So input

input.w = 2412
result1=input>>8
result2=input&$FF
MessageRequester("Results:",Str(result1)+" , "+Str(result2),0)
End

Ralf,

2412 decimal = 96c hex

Therefore:
result1 = 9 hex, 9 decimal
result2 = 6c hex, 108 decimal

Shifting by eight bits and masking the lower eight bits works for hex values, not decimal. If you want to get the decimal results, you must divide by 10^2 = 100, like this: Regards,
Eric

Thanks all.

ebs, your code is exactly what i needed but i m not sure if there could be problems by dividing by 100. Is there no problem that i get wrong result, even of the floats by dividing by 100. I m not sure if there could be any round up/down problems!?

Ralf,

The division by 100 should work for all decimal integer values, but not for floats.

Eric