Java 排除子目录的过滤器映射 url-pattern

Filter mapping url-pattern that excludes subdirectories

提问by Stefan Rasmusson

Is there any way to make a filtermapping not include subdirectories?

有什么方法可以使过滤器映射不包含子目录？

For example.

例如。

I have .xhtml files in my context root, and I also have a subfolder named "test" with files with the same extension. Is there any may to map a filter to the files in context root and not to the files in the "test" directory?

我的上下文根中有 .xhtml 文件，而且我还有一个名为“test”的子文件夹，其中包含具有相同扩展名的文件。是否可以将过滤器映射到上下文根中的文件而不是“test”目录中的文件？

采纳答案by BalusC

The url-patternis indeed restrictive in matching. It only allows exact, prefix or suffix matchnig. Not midst/overall/regex matching. So e.g. /*.xhtmlwhat you intend to do ain't going to work.

该url-pattern是匹配确实是限制。它只允许精确、前缀或后缀匹配。不是中间/整体/正则表达式匹配。因此，例如/*.xhtml，您打算做的事情是行不通的。

If you want to exclude XHTML in the /testfolder only, then your best is really a Filterlistening on an url-patternof *.xhtmlwhich does basically the following job in doFilter()method:

如果你只想在/test文件夹中排除 XHTML ，那么你最好的方法就是Filter监听其中的一个url-pattern，*.xhtml基本上在doFilter()方法中完成以下工作：

// First cast ServletRequest to HttpServletRequest.
HttpServletRequest hsr = (HttpServletRequest) request;

// Check if requested resource is not in /test folder.
if (!hsr.getServletPath().startsWith("/test/")) {
    // Not in /test folder. Do your thing here.
}

The HttpServletRequest#getServletPath()basically returns the part of the request URI from the context path on.

所述HttpServletRequest#getServletPath()基本上返回从上下文路径请求URI的一部分。

You can if necessary configure the value /testas an <init-param>of the filter so that you can control the value from inside the web.xmlinstead of in the Filter's code.

如有必要，您可以将该值配置/test为<init-param>过滤器的一个，以便您可以从web.xml过滤器代码内部而不是在过滤器代码中控制该值。

回答by Gopi

Some containers (like resin) define an extension to filter mapping definition in web.xml which allows specifying url-regexpwhich allows matching the urls based on regular expression. (reference). See if your container has something like this.

一些容器（如树脂）在 web.xml 中定义了过滤器映射定义的扩展，它允许指定url-regexp哪个允许基于正则表达式匹配 url。（参考）。看看你的容器有没有这样的东西。

回答by mdma

Using what is available for standard servlets/filters, there is no direct way to exclude the subdirectory, since *.extstyle mappings include subdirectories. You could work around this by declaring another filter that specifically handles the /test/*.xhtmlmapping and handles that accordingly.

使用可用于标准 servlet/过滤器的内容，没有直接的方法来排除子目录，因为*.ext样式映射包括子目录。您可以通过声明另一个专门处理/test/*.xhtml映射并相应处理的过滤器来解决此问题。

回答by softskin

The solution, basically you need your own Filter class and define excluding URLs as init-param for filter

解决方案，基本上你需要自己的 Filter 类并将排除 URL 定义为过滤器的 init-param

package test;

import java.io.IOException;
import java.util.ArrayList;
import java.util.List;

import javax.servlet.Filter;
import javax.servlet.FilterChain;
import javax.servlet.FilterConfig;
import javax.servlet.ServletException;
import javax.servlet.ServletRequest;
import javax.servlet.ServletResponse;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;

public class LoginValidationFilter implements Filter {

    private String loginPage = "";
    private List<String> excludedURLs;

    public void init(FilterConfig filterConfig) throws ServletException {
        this.loginPage = filterConfig.getInitParameter("loginPage");

        String[] excluded = filterConfig.getInitParameter("excludedURLs").split(";");
        excludedURLs = new ArrayList<String>();
        for (int i = 0; i < excluded.length; i++) {
            excludedURLs.add(excluded[i]);
        }
    }

    public void destroy() {
        this.loginPage = "";
    }

    public void doFilter(ServletRequest request, ServletResponse response, FilterChain filterChain) throws IOException,
            ServletException {

        HttpServletRequest httpRequest = request instanceof HttpServletRequest ? (HttpServletRequest) request : null;
        HttpServletResponse httpResponse = response instanceof HttpServletResponse ? (HttpServletResponse) response
                : null;

        if (httpRequest == null || httpResponse == null) {
            filterChain.doFilter(request, response);
            return;
        }

        boolean isExcludedURL = false;
        for (int i = 0; i < excludedURLs.size(); i++) {
            if (httpRequest.getRequestURL().indexOf(excludedURLs.get(i)) > -1) {
                isExcludedURL = true;
                break;
            }
        }

        if (isExcludedURL) {
            filterChain.doFilter(request, response);
        } else {
            if (UserUtil.validateUserLogin(httpRequest)) {
                filterChain.doFilter(request, response);
            } else {
                httpResponse.sendRedirect(httpRequest.getContextPath() + loginPage);
            }
        }
    }
}

And define the filter, mapping and exclude URLs in web.xml

并在 web.xml 中定义过滤器、映射和排除 URL

<filter>
    <filter-name>LoginValidationFilter</filter-name>
    <filter-class>test.LoginValidationFilter</filter-class>
    <init-param>
        <param-name>loginPage</param-name>
        <param-value>/login.jsp</param-value>
    </init-param>
    <init-param>
        <param-name>excludedURLs</param-name>
        <param-value>/admin/;/assets/;/content/;/css/;/js/;/login.jsp;/login.cmd;/logout.cmd;forgot_password.jsp;pageName=</param-value>
    </init-param>
</filter>
<filter-mapping>
    <filter-name>LoginValidationFilter</filter-name>
    <url-pattern>/*</url-pattern>
</filter-mapping>

回答by acaruci

You can add a url-mapping value for every file you have in the context root, therefore the filter will apply to those files but none to the files in the test folder (actually it will only apply to those specific files). The following worked for me

您可以为上下文根中的每个文件添加 url-mapping 值，因此过滤器将应用于这些文件，但不适用于测试文件夹中的文件（实际上它仅适用于那些特定文件）。以下对我有用

<url-pattern>/index.faces</url-pattern>