The SQL Server money
datatype internally is a 64-bit integer with an implied scale of 4 decimal places. To quote Books Online, it is accurate "to ten-thousandsth of a currency unit." It is, the rough equivalent of a decimal(19,4)
.
The reason for the scale of 4 rather than 2 is to maintain precision in the results of arithmetic. Your ordinary currency value has a scale of 2 (e.g. $3.27) Multiplication or division of two numbers scaled to 2 decimal places gives a results that is precise to 4 decimal places: 9.23 divided by 3.27 yields a result of 2.82262996941896 (approximately). You can carry the result to whatever accuracy (number of decimal places) you desire. However, the result is only precise to 4 decimal places (2.8226) as the original values were only precise to 2 decimal places. That measurement is precise to within 1/2 of the smallest unit specified (+/- 0.005).
But I digress.
As a result of a SQL Server money
value having an implied scale of 4, ADO.Net converts the value to a System.Decimal with a scale of 4. And since System.Decimal tracks scale, when you convert it to string, you get 4 decimal places.
To get fewer, you can
- Round it before conversion, using the appropriate
Decimal.Round()
overload, or
- Format it as desired (eg.
(3.27M).ToString("0.00") ;
.
Hope this helps.
This program:
namespace Sandbox
{
using System ;
class Program
{
static void Main( string[] args )
{
decimal pi = (decimal) Math.PI ;
string piText = pi.ToString("0.00");
Console.WriteLine("PI to 2 decimal places is {0} one way, and {1:0.00} another" , piText , pi ) ;
return;
}
}
}
Produces what you'd expect:
PI to 2 decimal places is 3.14 one way, and 3.14 another
Cheers,
N.