使用Jsoup防御XSS攻击 | 云醉潇度

一、前言

跨站脚本攻击(Cross Site Scripting)，为了不和层叠样式表(Cascading Style Sheets, CSS)的缩写混淆，故将跨站脚本攻击缩写为XSS。恶意攻击者往Web页面里插入恶意Script代码，当用户浏览该页之时，嵌入其中Web里面的Script代码会被执行，从而达到恶意攻击用户的目的。
使用Jsoup可以有效的过滤不安全的代码。Jsoup使用白名单的机制来预防XSS攻击，比如白名单中规定只允许<span> 标签的存在，那么其他标签都会被过滤掉。

二、常见的XSS攻击

比如页面的某个表单允许用户输入任意内容，当某个调皮的用户输入如下内容：

保存后，你会发现页面文字都变成了红色！

或者输入<script>for(var i=0;i<10;i++){alert("fuck you");}，保存后页面将弹窗10次！

三、引入Jsoup

使用Maven构建一个简单的Spring Boot项目，在pom中引入：

       <!-- https://mvnrepository.com/artifact/org.jsoup/jsoup -->
        <dependency>
            <groupId>org.jsoup</groupId>
            <artifactId>jsoup</artifactId>
            <version>1.13.1</version>
        </dependency>

四、JsoupUtil

创建一个JsoupUtil工具类：

/**
 * Xss过滤工具
 *
 */
public class JsoupUtil {
    private static final Whitelist whitelist = Whitelist.basicWithImages();
    /*
     * 配置过滤化参数,不对代码进行格式化
     */
    private static final Document.OutputSettings outputSettings = new Document.OutputSettings().prettyPrint(false);
    static {
        /*
         * 富文本编辑时一些样式是使用style来进行实现的 比如红色字体 style="color:red;" 所以需要给所有标签添加style属性
         */
        whitelist.addAttributes(":all", "style");
    }

    public static String clean(String content) {
        return Jsoup.clean(content, "", whitelist, outputSettings);
    }
}

这里采用的白名单为basicWithImages，Jsoup内置了几种常见的白名单供我们选择，如下表所示：

白名单对象	标签	说明
none	无	只保留标签内文本内容
simpleText	b,em,i,strong,u	简单的文本标签
basic	a,b,blockquote,br,cite,code,dd,dl,dt,em,i,li,ol,p,pre,q,small,span,strike,strong,sub,sup,u,ul	基本使用的标签
basicWithImages	basic 的基础上添加了 img 标签 及 img 标签的 src,align,alt,height,width,title 属性	基本使用的加上 img 标签
relaxed	a,b,blockquote,br,caption,cite,code,col,colgroup,dd,div,dl,dt,em,h1,h2,h3,h4,h5,h6,i,img,li,ol,p,pre,q,small,span,strike,strong,sub,sup,table,tbody,td,tfoot,th,thead,tr,u,ul	在 basicWithImages 的基础上又增加了一部分部分标签

五、XssHttpServletRequestWrapper

创建一个XssHttpServletRequestWrapper，同过重写getParameter()，getParameterValues()和getHeader()方法来过滤HTTP请求中参数包含的恶意字符：

public class XssHttpServletRequestWrapper extends HttpServletRequestWrapper {

    HttpServletRequest orgRequest = null;

    private boolean isIncludeRichText = false;

    /**
     * Constructs a request object wrapping the given request.
     *
     * @param request The request to wrap
     * @throws IllegalArgumentException if the request is null
     */
    public XssHttpServletRequestWrapper(HttpServletRequest request) {
        super(request);
        orgRequest = request;
        this.isIncludeRichText = isIncludeRichText;
    }

    /**
     * 覆盖getParameter方法，将参数名和参数值都做xss过滤如果需要获得原始的值，则通过super.getParameterValues(name)来获取
     * getParameterNames,getParameterValues和getParameterMap也可能需要覆盖
     */
    @Override
    public String getParameter(String name) {
        if (("content".equals(name) || name.endsWith("WithHtml")) && !isIncludeRichText) {
            return super.getParameter(name);
        }
        name = JsoupUtil.clean(name);
        String value = super.getParameter(name);
        if (StringUtils.isNotBlank(value)) {
            value = JsoupUtil.clean(value);
        }
        return value;
    }

    @Override
    public String[] getParameterValues(String name) {
        String[] arr = super.getParameterValues(name);
        if (arr != null) {
            for (int i = 0; i < arr.length; i++) {
                arr[i] = JsoupUtil.clean(arr[i]);
            }
        }
        return arr;
    }

    /**
     * 覆盖getHeader方法，将参数名和参数值都做xss过滤如果需要获得原始的值，则通过super.getHeaders(name)来获取
     * getHeaderNames 也可能需要覆盖
     */
    @Override
    public String getHeader(String name) {
        name = JsoupUtil.clean(name);
        String value = super.getHeader(name);
        if (StringUtils.isNotBlank(value)) {
            value = JsoupUtil.clean(value);
        }
        return value;
    }

    /**
     * 获取原始的request
     */
    public HttpServletRequest getOrgRequest() {
        return orgRequest;
    }

    /**
     * 获取原始的request的静态方法
     */
    public static HttpServletRequest getOrgRequest(HttpServletRequest req) {
        if (req instanceof XssHttpServletRequestWrapper) {
            return ((XssHttpServletRequestWrapper) req).getOrgRequest();
        }
        return req;
    }
}

六、XssFilter

创建XssFilter，同过使用上面定义的XssHttpServletRequestWrapper类中的getParameter()等方法来保证参数得到了过滤：

@Slf4j
public class XssFilter implements Filter {

    // 是否过滤富文本内容
    private static boolean IS_INCLUDE_RICH_TEXT = false;

    public List<String> excludes = new ArrayList<String>();

    @Override
    public void init(FilterConfig filterConfig) throws ServletException {
        log.info("------------ xss filter init ------------");
        String isIncludeRichText = filterConfig.getInitParameter("isIncludeRichText");
        if (StringUtils.isNotBlank(isIncludeRichText)) {
            IS_INCLUDE_RICH_TEXT = BooleanUtils.toBoolean(isIncludeRichText);
        }
        String temp = filterConfig.getInitParameter("excludes");
        if (temp != null) {
            String[] url = temp.split(",");
            for (int i = 0; url != null && i < url.length; i++) {
                excludes.add(url[i]);
            }
        }
    }

    @Override
    public void doFilter(ServletRequest request, ServletResponse response, FilterChain chain) throws IOException, ServletException {
        log.info("------------ xss filter doFilter ------------");
        HttpServletRequest req = (HttpServletRequest) request;
        HttpServletResponse resp = (HttpServletResponse) response;
        if (handleExcludeURL(req, resp)) {
            chain.doFilter(request, response);
            return;
        }
        XssHttpServletRequestWrapper xssRequest = new XssHttpServletRequestWrapper((HttpServletRequest) request);
        chain.doFilter(xssRequest, response);
    }

    @Override
    public void destroy() {
        log.info("------------ xss filter destroy ------------");

    }

    private boolean handleExcludeURL(HttpServletRequest request, HttpServletResponse response) {
        if (excludes == null || excludes.isEmpty()) {
            return false;
        }
        String url = request.getServletPath();
        for (String pattern : excludes) {
            Pattern p = Pattern.compile("^" + pattern);
            Matcher m = p.matcher(url);
            if (m.find())
                return true;
        }
        return false;
    }
}

七、Spring Boot中配置XssFilter

使用JavaConfig的形式配置：

@Configuration
public class WebConfig {
    @Bean
    public FilterRegistrationBean xssFilterRegistrationBean() {
        FilterRegistrationBean filterRegistrationBean = new FilterRegistrationBean();
        filterRegistrationBean.setFilter(new XssFilter());
        filterRegistrationBean.setOrder(1);
        filterRegistrationBean.setEnabled(true);
        filterRegistrationBean.addUrlPatterns("/*");
        Map<String, String> initParameters = new HashMap<String, String>();
        initParameters.put("excludes", "/favicon.ico,/img/*,/js/*,/css/*");
        initParameters.put("isIncludeRichText", "true");
        filterRegistrationBean.setInitParameters(initParameters);
        return filterRegistrationBean;
    }
}