截取指定长度字符串

[\u0391-\uFFE5]匹配双字节字符（汉字+符号）

[\u4e00-\u9fa5]注意只匹配汉字，不匹配双字节字符

/// <summary>
/// 截取指定长度的字符串（Unicode占两位）
/// </summary>
/// <param name="source">指定字符串</param>
/// <param name="length">指定长度</param>
/// <returns>返回指定字符串长度的</returns>
public static string Sub(string source, int length)
{
    string temp = source;
    if (Regex.Replace(temp, "[\u0391-\uffe5]", "  ", RegexOptions.IgnoreCase).Length <= length)
    {
        return temp;
    }
    for (int i = temp.Length; i >= 0; i--)
    {
        temp = temp.Substring(0, i);
        if (Regex.Replace(temp, "[\u0391-\uffe5]", "  ", RegexOptions.IgnoreCase).Length <= length)
        {
            return temp + "";
        }
    }
    return "";
}

 上面这种执行效率比下面这种方法要高，耗时要短：

 

/// <summary>
/// 截取指定长度的字符串（Unicode占两位）
/// </summary>
/// <param name="source">指定字符串</param>
/// <param name="length">指定长度</param>
/// <returns>返回指定字符串长度的</returns>
public static string SLeft(string source, int length)
{
    Regex regex = new Regex("[\u0391-\uffe5]+", RegexOptions.Compiled);
    char[] stringChar = source.ToCharArray();
    string sb = "";
    int nLength = 0;
    for (int i = 0; i < stringChar.Length; i++)
    {
        if (nLength >= length)
        {
            break;
        }
        if (regex.IsMatch((stringChar[i]).ToString()))
        {
            sb += stringChar[i];
            nLength += 2;
        }
        else
        {
            sb += stringChar[i];
            nLength = nLength + 1;
        }
    }
    return sb.ToString();
}

 下面这种方法耗时最短，效率最高：

 

public static string GetStr(string source, int length)
{
    int count = 0;
    for (int i = 0; i < source.Length; i++)
    {
        if (((int)source[i]).ToString("X2").Length == 4)
            count += 2;
        else
            count += 1;
    }
    if (count > length)
    {
        StringBuilder sb = new StringBuilder();
        count = 0;
        for (int i = 0; i < source.Length; i++)
        {
            char temp = source[i];
            if (((int)temp).ToString("X2").Length == 4)
            {
                count += 2;
            }
            else
            {
                count += 1;
            }
            if (count < length)
            {
                sb.Append(temp);
            }
            else if (count == length)
            {
                sb.Append(temp);
                break;
            }
            else if (count > length)
            {
                sb.Append("");
                break;
            }
        }
        return sb.ToString();
    }
    else
    {
        return source;
    }
}

 

 

测试代码如下：

string source = "《中国》按时打发按时打发撒地方啊飒飒asdfasdfasd的法师打发似的<发（生的发生）地方按时打发是asdfasdfas";
int length = 10;

DateTime dt = DateTime.Now;
for (int i = 0; i < 1000; i++)
{
    string temp = StrFun.SLeft(source, length);
}
TimeSpan ts = DateTime.Now - dt;
Console.WriteLine(ts.TotalMilliseconds);


dt = DateTime.Now;
for (int i = 0; i < 1000; i++)
{
    string temp = StrFun.Sub(source, length);
}
ts = DateTime.Now - dt;
Console.WriteLine(ts.TotalMilliseconds);

dt = DateTime.Now;
for (int i = 0; i < 1000; i++)
{
    string temp = StrFun.GetStr(source, length);
}
ts = DateTime.Now - dt;
Console.WriteLine(ts.TotalMilliseconds);

Console.Read();

  

结果为：
2328.125
1515.625
15.625