按字节长度截取字符串(支持截取带HTML代码样式的字符串)

        /// <summary>
        /// 按字节长度截取字符串(支持截取带HTML代码样式的字符串)
        /// </summary>
        /// <param name="param">将要截取的字符串参数</param>
        /// <param name="length">截取的字节长度</param>
        /// <param name="end">字符串末尾补上的字符串</param>
        /// <returns>返回截取后的字符串</returns>
        public string SubstringToHTML(string param, int length, string end)
        {
            string Pattern = null;
            MatchCollection m = null;
            StringBuilder result = new StringBuilder();
            int n = 0;
            char temp;
            bool isCode = false; //是不是HTML代码
            bool isHTML = false; //是不是HTML特殊字符,如&nbsp;
            char[] pchar = param.ToCharArray();
            for (int i = 0; i < pchar.Length; i++)
            {
                temp = pchar[i];
                if (temp == '<')
                {
                    isCode = true;
                }
                else if (temp == '&')
                {
                    isHTML = true;
                }
                else if (temp == '>' && isCode)
                {
                    n = n - 1;
                    isCode = false;
                }
                else if (temp == ';' && isHTML)
                {
                    isHTML = false;
                }

                if (!isCode && !isHTML)
                {
                    n = n + 1;
                    //UNICODE码字符占两个字节
                    if (System.Text.Encoding.Default.GetBytes(temp + "").Length > 1)
                    {
                        n = n + 1;
                    }
                }

                result.Append(temp);
                if (n >= length)
                {
                    break;
                }
            }
            result.Append(end);
            //取出截取字符串中的HTML标记
            string temp_result = result.ToString().Replace("(>)[^<>]*(<?)", "$1$2");
            //去掉不需要结素标记的HTML标记
            temp_result = temp_result.Replace(@"</?(AREA|BASE|BASEFONT|BODY|BR|COL|COLGROUP|DD|DT|FRAME|HEAD|HR|HTML|IMG|INPUT|ISINDEX|LI|LINK|META|OPTION|P|PARAM|TBODY|TD|TFOOT|TH|THEAD|TR|area|base|basefont|body|br|col|colgroup|dd|dt|frame|head|hr|html|img|input|isindex|li|link|meta|option|p|param|tbody|td|tfoot|th|thead|tr)[^<>]*/?>",
             "");
            //去掉成对的HTML标记
            temp_result = temp_result.Replace(@"<([a-zA-Z]+)[^<>]*>(.*?)</\1>", "$2");
            //用正则表达式取出标记
            Pattern = ("<([a-zA-Z]+)[^<>]*>");
            m = Regex.Matches(temp_result, Pattern);

            ArrayList endHTML = new ArrayList();

            foreach (Match mt in m)
            {
                endHTML.Add(mt.Result("$1"));
            }
            //补全不成对的HTML标记
            for (int i = endHTML.Count - 1; i >= 0; i--)
            {
                result.Append("</");
                result.Append(endHTML[i]);
                result.Append(">");
            }

            return result.ToString();
        }

转载于:https://www.cnblogs.com/xsmhero/archive/2009/06/16/1504443.html