[原创]正则匹配

var strHref = top.document.location.href;
var regu = "OriginalPic";
var re = new RegExp(regu);
    if (strHref.search(re) != -1) {
       “have”;
    }

以上是JS的包含正则 ，search 确定一个字符串是否包含另外一段字符

 

 

------------------------------------------------------------------

C#的正则表达式

 

html = "5800XM报价5480</a> <span>2008-11-11 </span></li><li><i>[<a href=\"http://yn.it168.com\">云南</a>]</i> <a title=\"[昆明]昆明十一月份最受关注十大手机\" href=\"http://publish.it168.com/2008/1130/20081130005201.shtml\">[昆明]";


 string zhengze = "<div class=\"tit9\">(.+?)</div>";
Match m = Regex.Match(html, zhengze);  //正则匹配 
if (!string.IsNullOrEmpty(m.Groups[1].Value)){}

 

 去图片的正则也可以简化一下 <img\s[^<>]*?>

过滤所有HTML标签

 

        html = System.Text.RegularExpressions.Regex.Replace(html, "<[^>]*>", "", RegexOptions.IgnoreCase); 
            html = System.Text.RegularExpressions.Regex.Replace(html, "<[^>]*>", "", RegexOptions.IgnoreCase);  
            return html;

 

 

 过滤所有换行

 

        /// <summary>
        /// 别换行 
        /// </summary>
        /// <param name="html"></param>
        /// <returns></returns>
        private static string NonrBr(string html)
        { 
            html = Regex.Replace(html, @"[\n\r]", "");
            html = html.TrimEnd((char[])"\n\r".ToCharArray());
            return html;
        }

 

 

 

还有一些常用的处理

 

  private static void regexReplaceFilter(ref string answerStr)
        {
            MatchCollection mc = null;
            string zhengZe = "";
            //去链接
            zhengZe = "<a[^>]*?href=[^>]*?>";
            mc = Regex.Matches(answerStr, zhengZe);
            foreach (Match var in mc)
            {
                answerStr = answerStr.Replace(var.Value, "");
            }
            //去图片
            zhengZe = @"<img\b[^<>]*?\bsrc[\s\t\r\n]*=[\s\t\r\n]*[""']?[\s\t\r\n]*(?<imgUrl>[^\s\t\r\n""'<>]*)[^<>]*?/?[\s\t\r\n]*>";
            mc = Regex.Matches(answerStr, zhengZe);
            foreach (Match var in mc)
            {
                answerStr = answerStr.Replace(var.Value, "");
            }
            //去网址
            zhengZe = @"http:\/\/[A-Za-z0-9]+\.[A-Za-z0-9]+[\/=\?%\-&_~`@[\]\':+!]*([^<>])*";
            mc = Regex.Matches(answerStr, zhengZe);
            foreach (Match var in mc)
            {
                answerStr = answerStr.Replace(var.Value, "");
            }
            //去QQ
            zhengZe = @"[1-9]\d{4,8}";
            mc = Regex.Matches(answerStr, zhengZe);
            foreach (Match var in mc)
            {
                answerStr = answerStr.Replace(var.Value, "");
            }
            //去电话  
            zhengZe = @"((\(\d{3}\))|(\d{3}\-))?(\(0\d{2,3}\)|0\d{2,3}-)?[1-9]\d{6,7}";
            mc = Regex.Matches(answerStr, zhengZe);
            foreach (Match var in mc)
            {
                answerStr = answerStr.Replace(var.Value, "");
            }

            //去手机 
            zhengZe = @"((\(\d{3}\))|(\d{3}\-))?13\d{9}";
            mc = Regex.Matches(answerStr, zhengZe);
            foreach (Match var in mc)
            {
                answerStr = answerStr.Replace(var.Value, "");
            }
        }