过滤在线编辑器产生的不安全html代码
<?php
/**
* 过滤在线编辑器产生的不安全html代码.
*
* PHP versions 4 and 5
*
* @copyright 版权所无,任意传播.
* @link http://www.52sunny.net
* @name html过滤
* @version v 0.0.10
* @author Lucklrj (sunny_lrj@yeah.net,qq:7691272)
* @lastmodified 2006-06-09 10:42 (Tue, 2006-06-09)
* @notice 此版本只过滤js,框架,表单。
作者能力有限,使用本程序若产生任何安全问题,与本人无关。
欢迎来信与我交流。
*/
$str="<tr><td bgcolor=''#FFFFFF''>
<div style=''url(123.offsetWidth)>";
//$str="url(javascript:x)";
/*不需要过滤的数组*/
$htm_on=array(
"<acronym","acronym>",
"<baseFont","baseFont>",
"<button","button>",
"<caption","caption>",
"<clientInformation","clientInformation>",
"<font","font>",
"<implementation","implementation>",
"<button","button>",
"<location","location>",
"<option","option>",
"<selection","selection>",
"<strong","strong>");
$htm_on_uper=array(
"<ACRONYM","span>ACRONYM>",
"<BASEFONT","BASEFONT>",
"<BUTTON","BUTTON>",
"<CAPTION","CAPTION>",
"<CLIENTINFORMATION","CLIENTINFORMATION>",
"<FONT","FONT>",
"<IMPLEMENTATION","IMPLEMENTATION>",
"<BUTTON","BUTTON>",
"<LOCATION","LOCATION>",
"<OPTION","OPTION>",
"<SELECTION","SELECTION>",
"<STRONG","STRONG>");
/*字符格式*/
$str=strtolower($str);
$str=preg_replace("/s+/", " ", $str);//过滤回车
$str=preg_replace("/ +/"an style="COLOR: #000000">, " ", $str);//过滤多个空格
/*过滤/替换几种形式的js*/
$str=preg_replace("/<(script.*?)>(.*?)<(/script.*?)>/si","",$str);//删除<script>。。。</script>格式,
//$str=preg_replace("/<(script.*?)>(.*?)<(/script.*?)>/si","<\1>\2<\3>",$str);//替换为可以显示的,
$str=preg_replace("/<(script.*?)>/si","",$str);//删除<script>未封闭
//$str=preg_replace("/<(script.*?)>/si","<\1>",$str);//替换未封闭
/*删除/替换表单*/
$str=preg_replace("/<(/?form.*?)>/si","",$str);//删除表单
//$str=preg_replace("/<(/?form.*?)>/si","<\1>",$str);//替换表单
$str=preg_replace("/<(i?frame.*?)>(.*?)<(/i?frame.*?)>/si","",$str);//删除框架
//$str=preg_replace("/<(i?frame.*?)>(.*?)<(/i?frame.*?)>/si","<\1>\2<\3>",$str);//替换框架
/*过滤on事件*/
$str=preg_replace("/href=(.+?)(["|''| |>])/ie","''href=''.strtoupper(''\1'').''\2''",$str);//把href=涉及到的on转换为大写。
$str=str_replace($htm_on,$htm_on_uper,$str);//把<font,font>换为大写, dhtml标签字符,正则判断太烦琐,采用转换办法。
$str=preg_replace("/(on[^ .<>]+?)([ |>])/s","\2",$str);//取掉on事件
/*过滤超级连接的js*/
$str=preg_replace("/(href|src|background|url|dynsrc|expression|codebase)[=:(]([ "'']*?w+..*?|javascript|vbscript:[^>]*?)()?)([ >/])/si","\1=''#'' \3\4",$str);//取掉href=javascript:
//返回小写字符
$str=strtolower($str);
$str=str_replace("&","&",$str);
echo $str;
?>