C#实现的html内容截取

using System.Collections.Generic;
public class Tool{
    /// <summary>
    /// 删除标签
    /// </summary>
    /// <param name="tags"></param>
    /// <param name="tag"></param>
    private static void RemoveTag(List<string> tags, string tag)
    {
        for (int i = tags.Count - 1; i >= 0; i--) if (tags[i].IndexOf(tag) == 1) { tags.RemoveAt(i); break; }
    }
    /// <summary>
    /// 按文本内容长度截取HTML字符串(支持截取带HTML代码样式的字符串)
    /// </summary>
    /// <param name="html">将要截取的字符串参数</param>
    /// <param name="len">截取的字节长度</param>
    /// <param name="endString">字符串末尾补上的字符串</param>
    /// <returns>返回截取后的字符串</returns>
    public static string HTMLSubstring(string html, int len, string endString)
    {
            string r = "";
        if (!string.IsNullOrEmpty(html) && html.Length > len)
        {
            MatchCollection mcentiry, mchtmlTag;
            List<string> inputHTMLTag = new List<string>();
            string tmpValue, nowtag, losetag;
            int rWordCount = 0, wordNum = 0, i = 0;
            Regex rxSingle = new Regex("^<(br|hr|img|input|param|meta|link|wbr)", RegexOptions.Compiled | RegexOptions.IgnoreCase)//是否单标签正则
                , rxEndTag = new Regex("</[^>]+>", RegexOptions.Compiled)//是否结束标签正则
                , rxTagName = new Regex("</?([a-z\\d]+)[^>]*>", RegexOptions.Compiled | RegexOptions.IgnoreCase)//获取标签名正则
                , rxHtmlTag = new Regex("<[^>]+>", RegexOptions.Compiled)//html标签正则
                , rxEntity = new Regex("&[a-z]{1,9};", RegexOptions.Compiled | RegexOptions.IgnoreCase)//实体正则
                , rxEntityReverse = new Regex("§", RegexOptions.Compiled)//反向替换实体正则
                ;
            html = html.Replace("§", "&#167;");//替换字符§为他的实体“&#167;”，以便进行下一步替换
            mcentiry = rxEntity.Matches(html);//收集实体对象到匹配数组中
            html = rxEntity.Replace(html, "§");//替换实体为特殊字符§，这样好控制一个实体占用一个字符
            mchtmlTag = rxHtmlTag.Matches(html);//收集html标签到匹配数组中

            string[] arrWord = rxHtmlTag.Split(html);//拆分
            wordNum = arrWord.Length;


            //获取指定内容长度及HTML标签
            for (; i < wordNum; i++)
            {
                if (rWordCount + arrWord[i].Length >= len) r += arrWord[i].Substring(0, len - rWordCount) + endString;
                else r += arrWord[i];


                rWordCount += arrWord[i].Length;//计算已经获取到的字符长度

                if (rWordCount >= len) break;

                //搜集已经添加的非单标签，以便封闭HTML标签对
                if (i < wordNum - 1)
                {
                    tmpValue = mchtmlTag[i].Value.ToLower();
                    if (!rxSingle.IsMatch(tmpValue))
                    { //不是单标签
                        if (rxEndTag.IsMatch(tmpValue) && inputHTMLTag.Count > 0)
                        {
                            nowtag = rxTagName.Match(tmpValue).Groups[1].Value.ToLower();
                            losetag = rxTagName.Match(inputHTMLTag[inputHTMLTag.Count - 1]).Groups[1].Value.ToLower();
                            inputHTMLTag.RemoveAt(inputHTMLTag.Count - 1);
                            if (nowtag != losetag)
                            {
                                RemoveTag(inputHTMLTag, nowtag);
                                r += "</" + losetag + ">";
                            }
                        }
                        else inputHTMLTag.Add(tmpValue);
                    }
                    r += tmpValue;
                }
            }
            //替换回实体
            for (i = 0; i < mcentiry.Count; i++) r = rxEntityReverse.Replace(r, mcentiry[i].Value, 1);
            //封闭标签
            for (i = inputHTMLTag.Count - 1; i >= 0; i--) r += "</" + rxTagName.Match(inputHTMLTag[i].ToString()).Groups[1].Value + ">";
        }
        else r = html;

        return r;
   }
}