Is there a "Number" struct/class in .NET?

Question

I am attempting to store a variable length number that can have leading zeros as a part of that number.

Is there a class in the .NET framework capable of storing values like this without losing information about leading zeros, and could surpass the upper limit of a long?

I am currently storing them in a class like this, is there any way I could write this class better in the event there isn't some struct or class available in the BCL:

[Serializable]
public class Number
{
    public int[] Array { get; set; }
    public int Length { get { return Array.Length; } }

    public Number(string number)
    {
        Array = new int[number.Length];

        for (int i = 0; i < number.Length; i++)
        {
            Array[i] = Convert.ToInt32(number[i].ToString());
        }
    }

    public Number(int[] array)
    {
        Array = array;
    }

    public int ToInt()
    {
        return Convert.ToInt32(ToString());
    }

    public override string ToString()
    {
        StringBuilder sb = new StringBuilder(Array.Length);

        foreach (int i in Array)
            sb.Append(i);

        return sb.ToString();
    }
}

The ability to use this as a struct would be very useful, as would the ability to check equality easily.

Items in bold/italic are the requirements of such a class.

If you really want something that can exceed the upper limit of the `long` type, you're going to need to rethink your `ToInt` method. — Dan Tao, Jan 29 '10 at 22:12
Essentially yes, however I wanted to enforce at the class/struct level the inability to use non-numeric data, and to make it clear wherever it is used in code that this was the intention. — Brett Allen, Jan 29 '10 at 22:14
Good point Dan, however that was added to support some values that were stored as strings, and converted to int. Messy way of doing things definitely. +1 for pointing that out, need to add a comment warning about that in the code. — Brett Allen, Jan 29 '10 at 22:15
Why do you need the leading digits? What kind of "number" is this? I say "number" in quotes, because it sounds to me like what you're really storing is a code, like a SSN or zip-code or whatever, not actually a number. That it happens to only be allowed to have digits doesn't make it any less of a code. The normal way to store codes is through strings. — Lasse V. Karlsen, Jan 29 '10 at 22:23
It's not what you're looking for, but there is an internal class in the `System` namespace called `Number`. It throws `FormatException`'s on certain bad WCF service calls. I mention it in case someone else searches `.NET Number class` and ends up here. http://reflector.webtropy.com/default.aspx/4@0/4@0/untmp/DEVDIV_TFS/Dev10/Releases/RTMRel/ndp/clr/src/BCL/System/Number@cs/1305376/Number@cs — Tim Goodman, Oct 05 '12 at 18:40

score 2 · Accepted Answer · edited May 23 '17 at 10:24

2

I'd suggest taking a look at this question about big integers in C#. You could expand them for the leading zeros issue.

edited May 23 '17 at 10:24

Community

1
1

answered Jan 29 '10 at 22:10

Samuel Carrijo

17,449
12
49
59

Thank you! Can definitely use that since need to rely on 3.5 for time being, and can add the code to handle leading zeros myself. – Brett Allen Jan 29 '10 at 22:18

score 2 · Answer 2 · answered Jan 29 '10 at 22:18

I see a number of problems with your class that should be addressed with the redesign.

Your class represents a value and is mutable.
The constructor taking an int array parameter only copies the reference.
The value can be larger than long but is returned as an int from the ToInt() method.

The first thing I would do to shrink the size of this class is use the array of integers as a constant stream of bits which make up a binary representation of your number. This would take a lot more care in order to ensure correctness of any desired operations, but would save you a significant amount of space. If you do that, I would also store a variable to keep track of leading 0's, or perhaps a variable "total digits" semantics.

Sorry this isn't an answer to your question, but hopefully it will help you in redesigning this your class.

score 0 · Answer 3 · answered Jan 29 '10 at 22:05

0

The BigInteger in .Net 4.0 addresses your upper limit requirement.

I think you'd still need a class like this to handle your leading zeros requirement.

answered Jan 29 '10 at 22:05

Austin Salonen

49,173
15
109
139

I did notice that when searching to see if Int128 had been implemented in .NET yet when someone posted (then deleted) their comment about using long long. I suppose I could create a struct in .NET 4.0 to hold the BigNumber and a "length" value to determine what leading zeros are needed when padding. – Brett Allen Jan 29 '10 at 22:10

score 0 · Answer 4 · answered Jan 29 '10 at 22:19

Until BigInteger is openly available, I would create your own class/struct with one public string property to get/set the initial number (this will maintain your leading zeros). Then have a ReadOnly property that parses the string and returns an IntX instance.

Is there a "Number" struct/class in .NET?

4 Answers4