java提取HTML正文的正则表达式??

Question

正则表达式怎样与才能最大限度地把html正文提取出来,即去掉了广告,格式,图片,只剩下正文...

Answer 1

public static string StripHTML(string strHtml)
  {
   string [] aryReg ={
          @"<script[^>]*?>.*?</script>",
          @"<(\/\s*)?!?((\w+:)?\w+)(\w+(\s*=?\s*(([""'])(\\[""'tbnr]|[^\7])*?\7|\w+)|.{0})|\s)*?(\/\s*)?>",
          @"([\r\n])[\s]+",
          @"&(quot|#34);",
          @"&(amp|#38);",
          @"&(lt|#60);",
          @"&(gt|#62);", 
          @"&(nbsp|#160);", 
          @"&(iexcl|#161);",
          @"&(cent|#162);",
          @"&(pound|#163);",
          @"&(copy|#169);",
          @"&#(\d+);",
          @"-->",
          @"<!--.*\n"
         
         };
   string [] aryRep = {
           "",
           "",
           "",
           "\"",
           "&",
           "<",
           ">",
           " ",
           "\xa1",//chr(161),
           "\xa2",//chr(162),
           "\xa3",//chr(163),
           "\xa9",//chr(169),
           "",
           "\r\n",
           ""
          };
   string newReg =aryReg[0];
   string strOutput=strHtml;
   for(int i = 0;i<aryReg.Length;i++)
   {
    Regex regex = new Regex(aryReg[i],RegexOptions.IgnoreCase );
    strOutput = regex.Replace(strOutput,aryRep[i]);
   }
   strOutput.Replace("<","");
   strOutput.Replace(">","");
   strOutput.Replace("\r\n","");   return strOutput;
  }

Answer 2

private String removehtmltag(String htmlstr) throws UnsupportedEncodingException
{
htmlstr = htmlstr.replaceAll("\n", "<BR>");
Pattern pat = Pattern.compile("\\s*<.*?>\\s*", Pattern.DOTALL
| Pattern.MULTILINE | Pattern.CASE_INSENSITIVE);
Matcher m = pat.matcher(htmlstr);
String rs = m.replaceAll("   ");
rs = rs.replaceAll("&nbsp", "   ");
rs = rs.replaceAll("&lt;", "<");
rs = rs.replaceAll("&gt;", ">");
rs = rs.replaceAll(" ", "");
return rs;
}