I wrote this string extension awhile back, and I'm actually getting quite a bit of use out of it.

public static string Slice(this string str, int? start = null, int? end = null, int step = 1)
{
    if (step == 0) throw new ArgumentException("Step cannot be zero.", "step");

    if (start == null)
    {
        if (step > 0) start = 0;
        else start = str.Length - 1;
    }
    else if (start < 0)
    {
        if (start < -str.Length) start = 0;
        else start += str.Length;
    }
    else if (start > str.Length) start = str.Length;

    if (end == null)
    {
        if (step > 0) end = str.Length;
        else end = -1;
    }
    else if (end < 0)
    {
        if (end < -str.Length) end = 0;
        else end += str.Length;
    }
    else if (end > str.Length) end = str.Length;

    if (start == end || start < end && step < 0 || start > end && step > 0) return "";
    if (start < end && step == 1) return str.Substring((int)start, (int)(end - start));

    int length = (int)(((end - start) / (float)step) + 0.5f);
    var sb = new StringBuilder(length);
    for (int i = (int)start, j = 0; j < length; i += step, ++j)
        sb.Append(str[i]);
    return sb.ToString();
}

Since it's in all my projects now, I'm wondering if I could have done it better. More efficient, or would it produce unexpected results in any case?

Slice. It works like Python's array notation.

 "string"[start:end:step]

Many other languages have something like this too. string.Slice(1) is equivalent to string.Substring(1). string.Substring(1,-1) trims off the first and last character. string.Substring(null,null,-1) will reverse the string. string.Substring(step:2) will return a string with every other character... also similar to c# string slice

Edit
Report