好东西，老外用正则表达式写的HTML分离函数

发布：dxy 字体：[增加减小] 类型：转载

存成.asp文件，执行，你用ASPHTTP抓内容的时候用这个很爽，当然自己要改进一下了
<%
Option Explicit
Function stripHTML(strHTML)
'Strips the HTML tags from strHTML
  Dim objRegExp, strOutput
  Set objRegExp = New Regexp
  objRegExp.IgnoreCase = True
  objRegExp.Global = True
  objRegExp.Pattern = "<.+?>"
  'Replace all HTML tag matches with the empty string
  strOutput = objRegExp.Replace(strHTML, "")
  'Replace all < and > with < and >
  strOutput = Replace(strOutput, "<", "<")
  strOutput = Replace(strOutput, ">", ">")
  stripHTML = strOutput    'Return the value of strOutput
  Set objRegExp = Nothing
End Function

%>
<form method="post" id=form1 name=form1>
  <b>Enter an HTML String:</b><br>
  <textarea name="txtHTML" cols="50" rows="8" wrap="virtual"><%=Request("txtHTML")%></textarea>
  <p>
  <input type="submit" value="Strip HTML Tags!" id=submit1 name=submit1>
</form>
<% if Len(Request("txtHTML")) > 0 then %>
    <p><hr><p>
    <b><u>View of string <i>with no</i> HTML stripping:</u></b><br>
    <xmp>
    <%=Request("txtHTML")%>
    </xmp><p>
    <b><u>View of string <i>with</i> HTML stripping:</u></b><br>
    <pre>
    <%=StripHTML(Request("txtHTML"))%>
    </pre>
<% End If %>