Powerbasic port to Freebasic MOD operator

[rpkelly]

I'm just beginning my code port and addressing the differences. Here is my handling of the MOD differences.

Code: Select all

Function cmAMod(ByVal x as Double, ByVal y as Double) as Double

' Variation of x MOD y for Real Numbers adjusted so that the modulus
' of a multiple of the divisor is the divisor itself rather than zero.
' 
' If x MOD y = 0 then result is adjusted to y

    Function = x - y * (cmCeiling(x / y) - 1)

End Function
Function cmMod(ByVal x as Double, ByVal y as Double) as Double

' x MOD y for Real Numbers, y<>0

    Function = x - y * cmFloor(x / y)

End Function
Function cmCeiling(ByVal x as Double) as Long

' Return smallest interger greater than or equal to x

    Function = cmFloor(x * -1) * -1

End Function
Function cmFloor(ByVal x as Double) as Long

' Return largest integer less than or equal to x

    Function = Int(x)

End Function

[caseih]

It would help if you specified what differences you are addressing. Which things are different in FB and how do they behave in PB?

I certainly hope PB never had x mod y return y when x mod y is supposed to be 0! Because that is not a modulus by definition. If PB ever did that I would consider that a bug in the compiler.

[rpkelly]

FB MOD seems to only support integer types and PB supports real types.

[counting_pine]

I presume AMod's general purpose tends to be for finding padding amounts, e.g. tab sizes, where you want to round up to the next multiple, but always with a value of 1 or more.

[caseih]

I see. Just FYI there's a function in the C standard library called fmod() that you can use from FreeBASIC. It's in the "crt/math.bi" file.

Code: Select all

#include "crt/math.bi"

print fmod(3.3,1.2)

[rpkelly]

I presume AMod's general purpose tends to be for finding padding amounts, e.g. tab sizes, where you want to round up to the next multiple, but always with a value of 1 or more.

AMod is useful when a non zero result is wanted. For instance, assume you have an integer contains some number of months and you want to know the ending month. If nMonths = 12 then nMonths MOD 12 is zero but nMonths AMOD 12 is 12.

[frisian]

For the function cmCeiling:
replace Function = cmFloor(x * -1) * -1 (need's two multi ply's)
with Function = -cmFloor( -x) (same result but faster).