how to avoid floating number problems with org-table for finances?

how to avoid floating number problems with org-table for finances?

[Thomas Koch]

Hi,

I use a simple org-mode table to sum up the expenses and incomes of a small 
private project. Now I remembered that I should know as a programmer to NEVER 
EVER USE FLOATS FOR FINANCES!!!

But I did...

|                                        0.3 |
|                                        0.3 |
|                                        0.3 |
| 0.9000000000000000222044604925031308084726 |
#+TBLFM: @4$1=vsum(@1..@3);%.40f

Do you know any quirck how I can continue using point notation but be sure 
that the numbers are added correctly? I think I'd need another vsum method 
that would

- find the number of decimal points of the most precise field
- parse all numbers to integers by filling up zeros at the end and ignoring 
the decimal point
- sum up
- write the result with a decimal point at the right position

Best regards,

Thomas Koch, http://www.koch.ro

Re: how to avoid floating number problems with org-table for finances?

[Johan Sandblom]

Not what you asked for, but how about simply rounding it?

2012-10-12 19:06 skrev Thomas Koch:
> | 0.9000000000000000222044604925031308084726 |
> #+TBLFM: @4$1=vsum(@1..@3);%.40f

#+TBLFM: @4$1=round(vsum(@1..@3);%.40f

-- 
Johan Sandblom, MD PhD
m +46735521477
What is wanted is not the will to believe, but the
will to find out, which is the exact opposite
--Bertrand Russell

Re: how to avoid floating number problems with org-table for finances?

[Johan Sandblom]

Sorry, I meant

#+TBLFM: @4$1=round(vsum(@1..@3),2);%.40f

Johan

-- 
Johan Sandblom, MD PhD
m +46735521477
What is wanted is not the will to believe, but the
will to find out, which is the exact opposite
--Bertrand Russell

Re: how to avoid floating number problems with org-table for finances?

[Michael Brand]

Hi Thomas

I hope it’s not too late but I got into some backlog with reading the list.

On Fri, Oct 12, 2012 at 7:06 PM, Thomas Koch <thomas@koch.ro> wrote:
> #+TBLFM: @4$1=vsum(@1..@3);%.40f

For your example I would recommend:
#+TBLFM: @4$1=vsum(@1..@3); f-2
or if you really want 40 digits precision for the calculations behind:
#+TBLFM: @4$1=vsum(@1..@3); p40 f-2
Please read my explanations here:
http://orgmode.org/worg/org-faq.html#table-high-precision

The integer calculation you suggested with decimal point notation is
not possible out of the box and if at all I would rather go the way of
Calc Fractions to get infinite precision:
| 3:10 |
| 3:10 |
| 3:10 |
| 9:10 |
#+TBLFM: @4$1=vsum(@1..@3); F

Michael