>For those of you who need to perform calculations with precision better than 16 in VFP6 there is a new file YMORE16.exe to download from WWW.FOXITE.COM.
>
>YMORE16 works with positive and negative numbers written and stored as STRING.
>YMORE16 performs addition or subtraction on two numbers and returns the result as a STRING.
>Limits on number of decimals remain the same: maximum number of decimal places is 18 and the minimum is zero.
>
>Syntax examples:
>?ymore16("123456789012345.1234589", "+", "98765432109872345.1234589")
>?ymore16("+123456789012345.1234589", "", "-98765432109872345.1234589")
>?ymore16("-123456789012345.1234589", "-", "-98765432109872345.1234589")
>
>I will appreciate feedback with comments, opinions, bugs reports etc.
>
>Yuri Rubinov
>rubinov@juno.com

Yuri,

The basic idea is great - every now and then someone needs greater precision.

What does the second example do - the one with the empty operator?

I would suggest the following enhancements:

1) Extend the concept to multiplication and division. The algorithm is more complicated, of course, than addition or subtraction, and the calculations would be slow if you multiply, say, two 500-digit numbers. Ask me if you need details on the algorithm. The basic idea, of course, is to apply the multiplication algorithm you learned in primary school. Intermediate results have to be stored - and incremented - in arrays or something similar.

2) Instead of limiting to 16 decimals digits, you might provide an (optional) parameter for the number of digits in the result. Let's say that the default remains at 16 - this would seem a reasonable default to me. I mean, if you scan a text one byte at a time, there is no need to arbitrarily limit the input and the results to 16 decimals.

Hilmar.