我从Microsoft.VisualBasic.Strings中做了一个c#移植。我不知道他们为什么把这么有用的函数(从VB)保存在系统之外。框架中的字符串，但仍然在Microsoft.VisualBasic下。同样的场景用于财务函数(例如microsoft . visualbasic . finance . pmt())。

public static string StrReverse(this string expression)
{
    if ((expression == null))
        return "";

    int srcIndex;

    var length = expression.Length;
    if (length == 0)
        return "";

    //CONSIDER: Get System.String to add a surrogate aware Reverse method

    //Detect if there are any graphemes that need special handling
    for (srcIndex = 0; srcIndex <= length - 1; srcIndex++)
    {
        var ch = expression[srcIndex];
        var uc = char.GetUnicodeCategory(ch);
        if (uc == UnicodeCategory.Surrogate || uc == UnicodeCategory.NonSpacingMark || uc == UnicodeCategory.SpacingCombiningMark || uc == UnicodeCategory.EnclosingMark)
        {
            //Need to use special handling
            return InternalStrReverse(expression, srcIndex, length);
        }
    }

    var chars = expression.ToCharArray();
    Array.Reverse(chars);
    return new string(chars);
}

///<remarks>This routine handles reversing Strings containing graphemes
/// GRAPHEME: a text element that is displayed as a single character</remarks>
private static string InternalStrReverse(string expression, int srcIndex, int length)
{
    //This code can only be hit one time
    var sb = new StringBuilder(length) { Length = length };

    var textEnum = StringInfo.GetTextElementEnumerator(expression, srcIndex);

    //Init enumerator position
    if (!textEnum.MoveNext())
    {
        return "";
    }

    var lastSrcIndex = 0;
    var destIndex = length - 1;

    //Copy up the first surrogate found
    while (lastSrcIndex < srcIndex)
    {
        sb[destIndex] = expression[lastSrcIndex];
        destIndex -= 1;
        lastSrcIndex += 1;
    }

    //Now iterate through the text elements and copy them to the reversed string
    var nextSrcIndex = textEnum.ElementIndex;

    while (destIndex >= 0)
    {
        srcIndex = nextSrcIndex;

        //Move to next element
        nextSrcIndex = (textEnum.MoveNext()) ? textEnum.ElementIndex : length;
        lastSrcIndex = nextSrcIndex - 1;

        while (lastSrcIndex >= srcIndex)
        {
            sb[destIndex] = expression[lastSrcIndex];
            destIndex -= 1;
            lastSrcIndex -= 1;
        }
    }

    return sb.ToString();
}

好吧，为了“不要重复你自己”，我提供以下解决方案:

public string Reverse(string text)
{
   return Microsoft.VisualBasic.Strings.StrReverse(text);
}

我的理解是这个实现，在VB中默认可用。NET，正确地处理Unicode字符。

如果在面试中你被告知不能使用Array。反过来，我想这可能是最快的一个。它不会创建新的字符串，只迭代数组的一半以上(即O(n/2)次迭代)

    public static string ReverseString(string stringToReverse)
    {
        char[] charArray = stringToReverse.ToCharArray();
        int len = charArray.Length-1;
        int mid = len / 2;

        for (int i = 0; i < mid; i++)
        {
            char tmp = charArray[i];
            charArray[i] = charArray[len - i];
            charArray[len - i] = tmp;
        }
        return new string(charArray);
    }

看看维基百科的条目。它们实现了字符串。反向扩展法。这允许你编写这样的代码:

string s = "olleh";
s.Reverse();

他们还使用ToCharArray/Reverse组合，这是这个问题的其他答案所建议的。源代码如下所示:

public static string Reverse(this string input)
{
    char[] chars = input.ToCharArray();
    Array.Reverse(chars);
    return new String(chars);
}

因为我喜欢两个答案-一个是使用字符串。创建，因此高性能和低分配和另一个正确性-使用StringInfo类，我决定需要一种组合方法。这是最终的字符串反转方法:)

private static string ReverseString(string str)
    {
        return string.Create(str.Length, str, (chars, state) =>
        {
            var enumerator = StringInfo.GetTextElementEnumerator(state);
            var position = state.Length;
            while (enumerator.MoveNext())
            {
                var cluster = ((string)enumerator.Current).AsSpan();
                cluster.CopyTo(chars.Slice(position - cluster.Length));
                position -= cluster.Length;
            }
        });
    }

还有一种更好的方法，使用StringInfo类的方法，它通过只返回索引来跳过Enumerator的大量字符串分配。

private static string ReverseString(string str)
    {
        return string.Create(str.Length, str, (chars, state) =>
        {
            var position = 0;
            var indexes = StringInfo.ParseCombiningCharacters(state); // skips string creation
            var stateSpan = state.AsSpan();
            for (int len = indexes.Length, i = len - 1; i >= 0; i--)
            {
                var index = indexes[i];
                var spanLength = i == len - 1 ? state.Length - index : indexes[i + 1] - index;
                stateSpan.Slice(index, spanLength).CopyTo(chars.Slice(position));
                position += spanLength;
            }
        });
    }

与LINQ解决方案相比的一些基准测试:

String length 20:

LINQ                       Mean: 2,355.5 ns   Allocated: 1440 B
string.Create              Mean:   851.0 ns   Allocated:  720 B
string.Create with indexes Mean:   466.4 ns   Allocated:  168 B

String length 450:

LINQ                          Mean: 34.33 us   Allocated: 22.98 KB
string.Create                 Mean:   19.13 us   Allocated: 14.98 KB
string.Create with indexes    Mean:   10.32 us   Allocated: 2.69 KB

private static string Reverse(string str)
        {
            string revStr = string.Empty;
            for (int i = str.Length - 1; i >= 0; i--)
            {
                revStr += str[i].ToString();
            }
            return revStr;
        }

比上述方法快

private static string ReverseEx(string str)
        {
            char[] chrArray = str.ToCharArray();
            int len = chrArray.Length - 1;
            char rev = 'n';
            for (int i = 0; i <= len/2; i++)
            {
                rev = chrArray[i];
                chrArray[i] = chrArray[len - i];
                chrArray[len - i] = rev;
            }
            return new string(chrArray);
        }