asked on

Need help with a rounding problem

I need some code (formula) that can return an integer value based on these rules:

1. No rounding

value returned value
0.9 0
1.1 1
1.5 1
.5 0

2. < 1 rounding

value returned value
.9 0
1.1 1
1.5 2
.5 0

3. < 0.5 rounding

value returned value
.3 0
.5 1
1.5 2
1.2 1

Obviously the value can be any decimal number so I need a formula or three that can handle this kind of rounding.

ASKER CERTIFIED SOLUTION

geobul

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

esoftbg

1.
var
I: Integer;
D: Double;
begin
D := 0.9;
I := Round(Int(D));
Edit1.Text := IntToStr(I);
end;

2.
var
I: Integer;
D: Double;
begin
D := 0.9;
I := Round(D);
Edit1.Text := IntToStr(I);
end;

DragonSlayer

Also check out the floor and ceil functions in the maths unit

esoftbg

procedure TForm1.Button1Click(Sender: TObject);
var
I: Integer;
DA: array[0..3] of Double;
begin
DA[0] := 0.9;
DA[1] := 1.1;
DA[2] := 1.5;
DA[3] := 0.5;
for I := 0 to 3 do
ListBox1.Items.Add(FloatToStr(Floor(DA[I])));
end;

procedure TForm1.Button2Click(Sender: TObject);
var
I: Integer;
DA: array[0..3] of Double;
begin
DA[0] := 0.9;
DA[1] := 1.1;
DA[2] := 1.5;
DA[3] := 0.5;
for I := 0 to 3 do
begin
if (DA[I]<1) then
ListBox2.Items.Add(FloatToStr(Floor(DA[I])))
else
ListBox2.Items.Add(FloatToStr(Round(DA[I])));
end;
end;

procedure TForm1.Button3Click(Sender: TObject);
var
I: Integer;
DA: array[0..3] of Double;
begin
DA[0] := 0.3;
DA[1] := 0.5;
DA[2] := 1.5;
DA[3] := 1.2;
for I := 0 to 3 do
begin
if (DA[I]<0.5) then
ListBox3.Items.Add(FloatToStr(Floor(DA[I])))
else
ListBox3.Items.Add(FloatToStr(Round(DA[I]+0.00000001)));
end;
end;

esoftbg

Post
Button1, Button2, Button3,
ListBox1, ListBox2, ListBox3
on the Form1 to see the results from above code.

geobul

I'm surprised :-(

BTW in the accepted answer:
- The first one will behave unexpectedly with negative numbers, i.e. '-1.5' will become '-2' instead of '-1'.
- I can't say anythung about 2 because I haven't understood the requirement.
- IMHO the last one is wrong. That constant depends on the type of the real variable and its specific representation (OS, processor, etc.).

esoftbg

Hi Geo, you are right !
I did not think abot the negative values at all ....
Your solutions about [1.] and [3.] are elegant !

So, how we could solve I am graded in mistake ?
I think the page editor could redirect points to you.

SOLUTION

esoftbg

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

esoftbg

kretzschmar,
please redirect the points of this question to geobul,
because his solutions are correct and elegant.
Emil.

geobul

Hi Emo,

The questioner is the one who decides which answer fits his needs. It is most likely that he doesn't care about negative numbers at all. So, there is no problem with the grading. I just wanted to leave my thougths for eventual future readers of this question. Hope it's clear now.

BTW. Congratulations, Emo, for your outstanding work here at Ex-Ex. I don't feel alone anymore ;-)

rutledgj

ASKER

You are right. Negative numbers were not a concern.