How to find the decimal places?

[royxu]

Hello,

I have to ask a dummy question.

X=7.56, I want to get how many decimal places in X. Here it is 2.

What function should I use?

Thanks!

[TimP]

I'm not certain what you mean by this, but it seems unlikely there is a single existing function which would do what you have in mind. If X is a binary floating point value, you could use internal write, probably with G format, to convert it to a character string with the appropriate number of digits (6, including those before the decimal point, for single precision). Then count the non-zero digits after the decimal point.

[Steven_L_Intel1]

It is important to note that many decimal fractional values cannot be exactly represented in binary floating point. So the approach Tim suggests, which rounds the value to some number of digits, is about the only plausible approach. You have to decide how many extra digits to round to, keeping in mind the fixed precision of the data type (6 or 7 significant digits for single precision, 15 for double.)

[royxu]

If you can provide sample codes, that would be very helpful.

I am not 100% sure how to do it.

Thanks!

Roy

[greldak]

Ok its not often I use recursion and its not Tims method but anyway

program CountDecPlaces

implicit none

! Variables
integer CountDecimal

! Body of CountDecPlaces
write(*,*)CountDecimal(1.0,0)
write(*,*)CountDecimal(1.234,0)
write(*,*)CountDecimal(1.34,0)
write(*,*)CountDecimal(10.0,0)
write(*,*)CountDecimal(12.1,0)
write(*,*)CountDecimal(1000.0001,0)

end program CountDecPlaces

recursive integer function CountDecimal(Number,Iterations)

integer iterations
realNumber

if (Number .eq. int(number)) then
CountDecimal=Iterations
else
CountDecimal=CountDecimal(Number*10,Iterations+1)
end if

end function

[royxu]

I solved it. Thanks!