XSS之xssprotect

liuzidong

浏览: 2287964 次
性别:
来自: 成都

最近访客更多访客>>

joechl

njdccy

mushroom12

musicxujun

博主相关

博客

微博

相册

留言

关于我

文章分类

社区版块

存档分类

博客分类：

Java
J2EE

html 脚本 javascript java

参考资料
1 跨网站脚本 http://zh.wikipedia.org/wiki/XSS
2 http://code.google.com/p/xssprotect/
一跨网站脚本介绍
     跨网站脚本（Cross-site scripting，通常简称为XSS或跨站脚本或跨站脚本攻击）是一种网站应用程序的安全漏洞攻击，是代码注入的一种。它允许恶意用户将代码注入到网页上，其他用户在观看网页时就会受到影响。这类攻击通常包含了HTML以及用户端脚本语言。
XSS攻击通常指的是通过利用网页开发时留下的漏洞，通过巧妙的方法注入恶意指令代码到网页，使用户加载并执行攻击者恶意制造的网页程序。这些恶意网页程序通常是JavaScript，但实际上也可以包括Java， VBScript， ActiveX， Flash 或者甚至是普通的HTML。攻击成功后，攻击者可能得到包括但不限于更高的权限（如执行一些操作）、私密网页内容、会话和cookie等各种内容。
二常用的XSS攻击手段和目的
盗用 cookie ，获取敏感信息。
利用植入 Flash ，通过 crossdomain 权限设置进一步获取更高权限；或者利用Java等得到类似的操作。
利用 iframe、frame、XMLHttpRequest或上述Flash等方式，以（被攻击）用户的身份执行一些管理动作，或执行一些一般的如发微博、加好友、发私信等操作。
利用可被攻击的域受到其他域信任的特点，以受信任来源的身份请求一些平时不允许的操作，如进行不当的投票活动。
在访问量极大的一些页面上的XSS可以攻击一些小型网站，实现DDoS攻击的效果。
三漏洞的防御和利用
避免XSS的方法之一主要是将用户所提供的内容进行过滤，许多语言都有提供对HTML的过滤：

    PHP的htmlentities()或是htmlspecialchars()。
    Python的cgi.escape()。
    ASP的Server.HTMLEncode()。
    ASP.NET的Server.HtmlEncode()或功能更强的Microsoft Anti-Cross Site Scripting Library
    Java的xssprotect(Open Source Library)。
    Node.js的node-validator。
四 xssprotect
在Eclipse中通过svn检出项目源地址：http://xssprotect.googlecode.com/svn/trunk/如下图

通用使用方式：http://code.google.com/p/xssprotect/wiki/HowTouse

package com.xss.example;

import java.io.IOException;
import java.io.StringReader;
import java.io.StringWriter;

import com.blogspot.radialmind.html.HTMLParser;
import com.blogspot.radialmind.html.HandlingException;
import com.blogspot.radialmind.xss.XSSFilter;

public class XSSTest {

	public static void main(String[] args) {
		String html = "<html><head> <title> New Document </title> " +
				"<script type='text/javascript'>  alert('dddd');   </script> " +
				"</head> <body>" +
				"222 <iframe  src='www.google.com'/>  1111" +
				"<embed ></embed>" +
				"<link>ddd</link>" +
				"</body></html>";
		String v = protectAgainstXSS(html);
		System.out.println(v);

	}
	
	public static String protectAgainstXSS( String html ) {
	    StringReader reader = new StringReader( html );
	    StringWriter writer = new StringWriter();
	    String text = null;
	    try {
	        // Parse incoming string from the "html" variable
	        HTMLParser.process( reader, writer, new XSSFilter(), true );
	        // Return the parsed and cleaned up string
	        text =  writer.toString();
	    } catch (HandlingException e) {
	        // Handle the error here in accordance with your coding policies...
	    }finally{
	    	try {
				writer.close();
				reader.close();
			} catch (IOException e) {				
				e.printStackTrace();
			}	    	
	    }
	    return text;
	}
}

在它的源代码中有二个类需要关注下：
BaseTestCase.java

public abstract class BaseTestCase extends TestCase {
	protected void testExecute( String html, String result ) {
		StringReader reader = new StringReader( html );
		StringWriter writer = new StringWriter();
		
		try {
			HTMLParser.process( reader, writer, new XSSFilter(), true );
			String buffer = new String( writer.toString() );
			System.out.println( buffer );
			assertEquals( result, buffer );
		} catch (HandlingException e) {
			e.printStackTrace();
			fail( e.getMessage() );
		}
	}
}

XSSFilter.java

/**
 * Copyright 2008 Gerard Toonstra
 *
 * As an exception, this particular file 
 * in the project is public domain to allow totally
 * free derivations of this code.
 * 
 */

package com.blogspot.radialmind.xss;

import java.util.HashSet;
import java.util.Set;

import com.blogspot.radialmind.html.IHTMLFilter;

/**
 * Implementation of a relatively simple XSS filter. This implementation removes
 * dangerous tags and attributes from tags. It does not verify the validity of 
 * URL's (that may contain links to JavaScript for example). It does not remove all
 * event handlers that may still contain XSS vulnerabilities. 
 * 
 * Embedded objects are removed because those may contain XSS vulnerabilities in 
 * their own scripting language (ActionScript for Flash for example).
 * 
 * Feel free to derive your own implementation from this file.
 * 
 * @author gt
 *
 */
public class XSSFilter implements IHTMLFilter {
	
	private static final Set<String> FORBIDDEN_TAGS = new HashSet<String>();
	
	// The tags to be removed. Case insensitive of course.
	static {
		FORBIDDEN_TAGS.add( "script" );
		FORBIDDEN_TAGS.add( "embed" );
		FORBIDDEN_TAGS.add( "object" );
		FORBIDDEN_TAGS.add( "layer" );
		FORBIDDEN_TAGS.add( "style" );
		FORBIDDEN_TAGS.add( "meta" );
		FORBIDDEN_TAGS.add( "iframe" );
		FORBIDDEN_TAGS.add( "frame" );
		FORBIDDEN_TAGS.add( "link" );
		FORBIDDEN_TAGS.add( "import" );
		FORBIDDEN_TAGS.add( "xml" );
	}
	
	/**
	 * This function is called to determine if an attribute should be filtered or not.
	 * 
	 * @param tagName	The name of the tag the attribute belongs to
	 * @param attrName	The name of the attribute to be filtered
	 * @param attrValue	The value of the attribute
	 */
	public boolean filterAttribute(String tagName, String attrName, String attrValue) {
		if ( attrName.toLowerCase().startsWith( "on" )) {
			return true;
		}
		
		return isScriptedAttributeValue( attrValue );
	}

	/**
	 * 	This method is called to determine if a tag should be filtered
	 * 
	 * @param tagName	The name of the tag that was parsed
	 */
	public boolean filterTag(String tagName) {
		if ( FORBIDDEN_TAGS.contains( tagName )) {
			return true;
		}
		return false;
	}

	/**
	 * This method is called to modify attribute values, if required
	 * 
	 * @param tagName	The name of the tag the attribute belongs to
	 * @param attrName	The name of the attribute within the tag
	 * @param attrValue		The value of the attribute
	 */
	public String modifyAttributeValue(String tagName, String attrName, String attrValue) {
		return attrValue;
	}

	/**
	 * This method is called to be able to modify the text of a node.
	 * 
	 * @param tagName	The name of the tag where the text is part of.
	 * @param text		The value of the text within the tagnode (within <tag>...</tag>)
	 */
	public String modifyNodeText(String tagName, String text) {
		return text;
	}
	
	/**
	 * Private method that determines if an attribute value is scripted
	 * (potentially loaded with an XSS attack vector).
	 * 
	 * @param attrValue	The value of the attribute
	 * @return "true" if the attribute is scripted. "false" if not.
	 */
	private boolean isScriptedAttributeValue( String attrValue ) {
		attrValue = decode( attrValue );
		attrValue = attrValue.trim().toLowerCase();

		if ( attrValue.contains( "javascript:" )) {
			return true;
		}
		if ( attrValue.contains( "mocha:" )) {
			return true;
		}
		if ( attrValue.contains( "eval" )) {
			return true;
		}
		if ( attrValue.contains( "vbscript:" )) {
			return true;
		}
		if ( attrValue.contains( "livescript:" )) {
			return true;
		}
		if ( attrValue.contains( "expression(" )) {
			return true;
		}
		if ( attrValue.contains( "url(" )) {
			return true;
		}
		if ( attrValue.contains( "&{" )) {
			return true;
		}
		if ( attrValue.contains( "&#" )) {
			return true;
		}
		return false;
	}
	
	/**
	 * Private method to remove control characters from the value
	 * 
	 * @param value	The value being modified
	 * @return	The value free from control characters
	 */
	private String decode( String value ) {
		value = value.replace("\u0000", "" );
		value = value.replace("\u0001", "" );
		value = value.replace("\u0002", "" );
		value = value.replace("\u0003", "" );
		value = value.replace("\u0004", "" );
		value = value.replace("\u0005", "" );
		value = value.replace("\u0006", "" );
		value = value.replace("\u0007", "" );
		value = value.replace("\u0008", "" );
		value = value.replace("\u0009", "" );
		value = value.replace("\n", "" );
		value = value.replace("\u000B", "" );
		value = value.replace("\u000C", "" );
		value = value.replace("\r", "" );
		value = value.replace("\u000E", "" );
		value = value.replace("\u000F", "" );
		value = value.replace("\u0010", "" );
		value = value.replace("\u0011", "" );
		value = value.replace("\u0012", "" );
		value = value.replace("\u0013", "" );
		value = value.replace("\u0014", "" );
		value = value.replace("\u0015", "" );
		value = value.replace("\u0016", "" );
		value = value.replace("\u0017", "" );
		value = value.replace("\u0018", "" );
		value = value.replace("\u0019", "" );
		value = value.replace("\u001A", "" );
		value = value.replace("\u001B", "" );
		value = value.replace("\u001C", "" );
		value = value.replace("\u001D", "" );
		value = value.replace("\u001E", "" );
		value = value.replace("\u001F", "" );
		return value;
	}
}

通过这个过滤就知道它要做什么了