Question

I want to write a function that format int and decimal differently into string

I have this code:

and I want to rewrite it to generics:

    public static string FormatAsIntWithCommaSeperator(int value)
    {
        if (value == 0 || (value > -1 && value < 1))
            return "0";
        return String.Format("{0:#,###,###}", value);
    }

    public static string FormatAsDecimalWithCommaSeperator(decimal value)
    {
        return String.Format("{0:#,###,###.##}", value);
    }


    public static string FormatWithCommaSeperator<T>(T value) where T : struct
    {
        string formattedString = string.Empty;

        if (typeof(T) == typeof(int))
        {
            if ((int)value == 0 || (value > -1 && value < 1))
            return "0";

            formattedString = String.Format("{0:#,###,###}", value);
        }

        //some code...
    }

    /// <summary>
    /// If the number is an int - returned format is without decimal digits
    /// </summary>
    /// <param name="value"></param>
    /// <returns></returns>
    public static string FormatNumberTwoDecimalDigitOrInt(decimal value)
    {
        return (value == (int)value) ? FormatAsIntWithCommaSeperator(Convert.ToInt32(value)) : FormatAsDecimalWithCommaSeperator(value);
    }

How can i use T in the function body?

What syntax should I use?

Answer 1

You can use the TypeCode enum for switch:

switch (Type.GetTypeCode(typeof(T)))
{
    case TypeCode.Int32:
       ...
       break;
    case TypeCode.Decimal:
       ...
       break;
}

Since C# 7.0 you can use pattern matching:

switch (obj)
{
    case int i:
       ...
       break;
    case decimal d:
       ...
       break;
    case UserDefinedType u:
       ...
       break;
}

Beginning with C# 8.0 you can use switch expressions:

string result = obj switch {
    int i => $"Integer {i}",
    decimal d => $"Decimal {d}",
    UserDefinedType u => "User defined {u}",
    _ => "unexpected type"
};