Count digits after point

[swkevin08]

as the others have said, it has something to do with the approach.
but look at this. maybe it will help you!

ps: doenst work for all numbers! for example x,42300000000000000001

Edit: e was randomly generated

Code:

const
  e = 0.0000000001;

function CountDigitsAfterPoint(AValue: Extended): Byte;
var r: Extended;
begin
  Result := 0;
  r := Int(AValue);
  while ((Frac(AValue) <> 0) and ((AValue - r) > e)) do
  begin
    Inc(Result);
    AValue := AValue * 10;
    r := Int(AValue+e);
  end;
end;

[WojTec]

Thanks, for all values I used early, is working correctly.
BTW: why e has that value, can have other values?

[swkevin08]

e is just a small number close to zero. it doesn't matter if there's one zero more or one less.

[Sir Rufo]

Zitat:

Zitat von WojTec (Beitrag 1248005)

Thanks, for all values I used early, is working correctly.
BTW: why e has that value, can have other values?

Have a look at the documentation System.Math.SameValue and System.Math.CompareValue and read what is the meaning of Epsilon.

Another helpful article is http://en.wikipedia.org/wiki/Machine_epsilon

[Furtbichler]

Zitat:

Zitat von WojTec (Beitrag 1247995)

So, reason is not important. Important is problem.
But if you want to know: (1) result is required for further computation purposes and (2) for formating. :)

The reason is always important, especially when people don't seem to exactly know what they are talking about.

Showing the number of decimals is definitely not something to please the eye. It is showing the accuracy of your value, so a value of, say, 100 (metres) measured up to a mm is correctly shown as "100,000 m" whereas the same value with an accuracy of 1m is shown as "100m".