写在前面
本文场景： 希望在网关上实现security统一进行权限认证，后续的服务间交互不再进行权限认证。但是系统有两个类型的账号，一个是普通用户，一个是系统后台管理员，完全是两个类型，希望创建给两个不同的登陆入口分别给两个类型的账号登录使用。
想到的解决方法有两个：
- 网关上的
security想办法创建多个登陆入口（本文描述的方法）； - 网关上的
security只对用户进行权限验证，对管理员的请求放行，管理员在自己的模块上进行权限验证。
本文思想： security实现多登陆入口的方法是创建多个SecurityFilterChain，即创建多个继承WebSecurityConfigurerAdapter的配置类。但是直接创建多个SecurityFilterChain并不会直接生效，因为SecurityFilterChain默认是全局生效的，且只会执行第一个匹配到的SecurityFilterChain，所以在configure(HttpSecurity http)方法中需要添加requestMatcher配置，限制生效范围。
前面3节废话写过滤器，第4节Security多登陆入口配置直接描述Security的配置。
1. Tomcat中Filter过滤器的执行流程
-
StandardWrapperValve的invoke方法，构造了一个ApplicationFilterChain对象，并执行了doFilter方法调用过滤器。ApplicationFilterChain是javax.servlet.FilterChain过滤器链的实现，用于管理针对特定请求的一组过滤器的执行，当定义的过滤器全部执行完成后，下一步将开始执行Servlet本身。 -
doFilter方法判断了一下SecurityManager（Java安全管理器）是否开启，随后调用internalDoFilter方法，如果pos < n（过滤器还未全部遍历），则通过ApplicationFilterConfig获取或者创建过滤器实例，后调用过滤器的doFilter(ServletRequest request, ServletResponse response, FilterChain chain)方法，如果过滤器调用完成则开始执行Servlet（servlet.service(request, response)）。
private void internalDoFilter(ServletRequest request,
ServletResponse response)
throws IOException, ServletException {
// 如果还有未执行的过滤器的话，执行下一项过滤器
if (pos < n) {
ApplicationFilterConfig filterConfig = filters[pos++];
try {
// 取得或者创建过滤器
Filter filter = filterConfig.getFilter();
if (request.isAsyncSupported() && "false".equalsIgnoreCase(
filterConfig.getFilterDef().getAsyncSupported())) {
request.setAttribute(Globals.ASYNC_SUPPORTED_ATTR, Boolean.FALSE);
}
// Java安全管理器是否开启
if( Globals.IS_SECURITY_ENABLED ) {
final ServletRequest req = request;
final ServletResponse res = response;
Principal principal =
((HttpServletRequest) req).getUserPrincipal();
Object[] args = new Object[]{req, res, this};
SecurityUtil.doAsPrivilege ("doFilter", filter, classType, args, principal);
} else {
filter.doFilter(request, response, this);
}
} catch (IOException | ServletException | RuntimeException e) {
throw e;
} catch (Throwable e) {
e = ExceptionUtils.unwrapInvocationTargetException(e);
ExceptionUtils.handleThrowable(e);
throw new ServletException(sm.getString("filterChain.filter"), e);
}
// 退出当前方法
return;
}
.......
// 如果过滤器全部执行完成将会开始执行这里，忽略n多细节，执行servlet
servlet.service(request, response);
}
FilterChain将自身当做参数传递给过滤器，internalDoFilter并不会再调用和控制过滤器链，而接下来过滤器的遍历操作通过filter实例来实现，如果过滤器中调用chain.doFilter(request, response)将进行下一个过滤器的遍历操作，也即表示通过当前过滤器，如果没有调用该方法，过滤器链不会继续遍历，也不会执行servlet，即被过滤器拦截。
2. 常见过滤器类型
GenericFilterBean：**主要是可以把Filter的初始化参数自动地set到继承于GenericFilterBean类的Filter中去。**将web.xml中filter标签中的配置参数init-param项作为bean的属性，实际的过滤工作留给他的子类来完成。
OncePerRequestFilter：继承GenericFilterBean类，主要功能是保证过滤器在任何servlet容器上每个请求都只能执行一次。首先通过(skipDispatch(httpRequest) || shouldNotFilter(httpRequest))判断是否跳过当前过滤器的执行，直接执行下一个过滤器；然后通过hasAlreadyFilteredAttribute属性判断当前过滤器是否已经调用过，确保没有被调用过后执行doFilterInternal(httpRequest, httpResponse, filterChain)方法实际执行过滤操作。
@Override
public final void doFilter(ServletRequest request, ServletResponse response, FilterChain filterChain)
throws ServletException, IOException {
if (!(request instanceof HttpServletRequest) || !(response instanceof HttpServletResponse)) {
throw new ServletException("OncePerRequestFilter just supports HTTP requests");
}
HttpServletRequest httpRequest = (HttpServletRequest) request;
HttpServletResponse httpResponse = (HttpServletResponse) response;
String alreadyFilteredAttributeName = getAlreadyFilteredAttributeName();
boolean hasAlreadyFilteredAttribute = request.getAttribute(alreadyFilteredAttributeName) != null;
// 判断是否跳过当前过滤器的执行
if (skipDispatch(httpRequest) || shouldNotFilter(httpRequest)) {
// Proceed without invoking this filter...
filterChain.doFilter(request, response);
}
// 判断是否跳过当前过滤器是否已经执行过了
else if (hasAlreadyFilteredAttribute) {
if (DispatcherType.ERROR.equals(request.getDispatcherType())) {
doFilterNestedErrorDispatch(httpRequest, httpResponse, filterChain);
return;
}
// Proceed without invoking this filter...
filterChain.doFilter(request, response);
}
else {
// 设置alreadyFilteredAttributeName为已经执行过,True
request.setAttribute(alreadyFilteredAttributeName, Boolean.TRUE);
try {
// 执行过滤操作
doFilterInternal(httpRequest, httpResponse, filterChain);
}
finally {
// 删除此请求的“已过滤”请求属性
request.removeAttribute(alreadyFilteredAttributeName);
}
}
}
由此可见每个请求都只能执行一次是基于request的，如果请求被重定向，或者标注的属性被删除，该过滤器还是会被重复执行的。
DelegatingFilterProxy：将要实现的Filter功能注册到IOC容器的一个Bean，这样就可以和Spring IOC容器进行完美的融合。DelegatingFilterProxy是过滤器Bean的代理，配置DelegatingFilterProxy时在Spring应用程序上下文中指定目标bean的名称targetBeanName，然后根据Bean名称从Spring 容器中获取到的代理Filter的实现类delegate，最终过滤操作调用的是委派的类delegate。
// 最终实现过滤的方法
protected void invokeDelegate(
Filter delegate, ServletRequest request, ServletResponse response, FilterChain filterChain)
throws ServletException, IOException {
// 委派类执行
delegate.doFilter(request, response, filterChain);
}
3. Security的入口：FilterChainProxy
FilterChainProxy：在Security流程中，FilterChainProxy作为DelegatingFilterProxy的委派类被ApplicationFilterChain执行。同样，主要的过滤逻辑在doFilterInternal方法中执行。
private void doFilterInternal(ServletRequest request, ServletResponse response,
FilterChain chain) throws IOException, ServletException {
// 拦截不安全的请求
FirewalledRequest fwRequest = firewall
.getFirewalledRequest((HttpServletRequest) request);
HttpServletResponse fwResponse = firewall
.getFirewalledResponse((HttpServletResponse) response);
// 从SecurityFilterChain列表第一个匹配的SecurityFilterChain中取得Filter链
// SecurityFilterChain列表中可能有多个匹配，但是这里只取得第一个匹配的，所以针对一个请求不能够配置多个SecurityFilterChain
List<Filter> filters = getFilters(fwRequest);
// 取得的过滤器为空，直接通过执行
if (filters == null || filters.size() == 0) {
if (logger.isDebugEnabled()) {
logger.debug(UrlUtils.buildRequestUrl(fwRequest)
+ (filters == null ? " has no matching filters"
: " has an empty filter list"));
}
fwRequest.reset();
chain.doFilter(fwRequest, fwResponse);
return;
}
//创建过滤器链
VirtualFilterChain vfc = new VirtualFilterChain(fwRequest, chain, filters);
// 执行过滤，过滤逻辑类似于ApplicationFilterChain
vfc.doFilter(fwRequest, fwResponse);
}
getFilters步骤实现了根据Request获取过滤器列表的操作，查看源码可以发现getFilters方法实现的是遍历SecurityFilterChain列表，获取第一个满足条件的chain，取得他的过滤器列表。而SecurityFilterChain默认是针对所有url生效的，也就是说，第一个chain就会被匹配到，后面的chain永远不会生效，所以需要添加requestMatcher配置，限制生效范围。
private List<Filter> getFilters(HttpServletRequest request) {
for (SecurityFilterChain chain : filterChains) {
if (chain.matches(request)) {
return chain.getFilters();
}
}
return null;
}
4. Security多登陆入口配置
- 创建两个继承
WebSecurityConfigurerAdapter的配置类，其中管理员配置AdminWebSecurityConfiguration加上.requestMatcher(new SkipPathAntMatcher("/admin/**"))，限制其范围仅包括/admin/**的请求； - 管理员配置加上了范围限制，用户配置还是全局范围，所以需要在用户
UserWebSecurityConfiguration中需要将管理员的/admin/**加入白名单，否则会被用户部分登录拦截，建议使用@Order注解用来声明Security的执行顺序，值越小，优先级越高，越先被执行/初始化。 - 下方提供的配置稍有不同，我在用户配置中创建了一个
.requestMatcher(new SkipPathAntMatcher("/admin/**"))，他将匹配除/admin/**外的所有请求。
package com.nineya.user.config;
import com.nineya.user.filter.AdminLoginAuthorizationTokenFilter;
import com.nineya.user.filter.UserLoginAuthorizationTokenFilter;
import com.nineya.user.handler.LoginAuthenticationFailureHandler;
import com.nineya.user.handler.LoginAuthenticationSuccessHandler;
import com.nineya.user.handler.MessageAuthenticationEntryPoint;
import com.nineya.user.service.AdminDetailsServiceImpl;
import com.nineya.user.service.UserDetailsServiceImpl;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.core.annotation.Order;
import org.springframework.security.authentication.AuthenticationProvider;
import org.springframework.security.authentication.dao.DaoAuthenticationProvider;
import org.springframework.security.config.annotation.authentication.builders.AuthenticationManagerBuilder;
import org.springframework.security.config.annotation.web.builders.HttpSecurity;
import org.springframework.security.config.annotation.web.configuration.EnableWebSecurity;
import org.springframework.security.config.annotation.web.configuration.WebSecurityConfigurerAdapter;
import org.springframework.security.config.annotation.web.configurers.AbstractHttpConfigurer;
import org.springframework.security.config.http.SessionCreationPolicy;
import org.springframework.security.crypto.bcrypt.BCryptPasswordEncoder;
import org.springframework.security.crypto.password.PasswordEncoder;
import org.springframework.security.web.authentication.UsernamePasswordAuthenticationFilter;
import org.springframework.security.web.util.matcher.AntPathRequestMatcher;
import javax.annotation.Resource;
/**
* 配置用户登录信息
* @author 殇雪话诀别
* 2020/11/22
*/
@EnableWebSecurity
public class WebSecurityConfiguration {
@Resource
private PasswordEncoder passwordEncoder;
@Resource
private LoginAuthenticationSuccessHandler successHandler;
@Resource
private LoginAuthenticationFailureHandler failureHandler;
/**
* 请求网页未登录时抛出的异常信息
*/
@Resource
private MessageAuthenticationEntryPoint entryPoint;
@Configuration
@Order(1)
public class UserWebSecurityConfiguration extends WebSecurityConfigurerAdapter {
@Resource
private UserDetailsServiceImpl detailsService;
@Resource
private UserLoginAuthorizationTokenFilter tokenFilter;
private final String[] authWhitelist = new String[]{"/oauth/login", "/oauth/auth/login"};
// //认证管理器
// @Override
// @Bean
// public AuthenticationManager authenticationManagerBean() throws Exception {
// return super.authenticationManagerBean();
// }
@Override
protected void configure(AuthenticationManagerBuilder auth) throws Exception {
auth
.userDetailsService(detailsService)
.passwordEncoder(passwordEncoder);
}
@Override
protected void configure(HttpSecurity http) throws Exception {
http
.requestMatcher(new SkipPathAntMatcher("/admin/**"))
.csrf(AbstractHttpConfigurer::disable)
.httpBasic(AbstractHttpConfigurer::disable)
.formLogin(n -> n
.loginPage("/oauth/login")
.loginProcessingUrl("/oauth/auth/login")
.successHandler(successHandler)
.failureHandler(failureHandler))
.logout(n -> n.logoutUrl("/oauth/logout"))
.exceptionHandling(n->n.authenticationEntryPoint(entryPoint))
.sessionManagement(n->n.sessionCreationPolicy(SessionCreationPolicy.STATELESS))
.addFilterBefore(tokenFilter, UsernamePasswordAuthenticationFilter.class)
.authorizeRequests(n -> n
.antMatchers(authWhitelist).permitAll()
.antMatchers("/user/api/**").authenticated()
.anyRequest().denyAll());
}
/**
* 用户的登录处理器
* @return 登录处理器
*/
public AuthenticationProvider userAuthenticationProvider() {
DaoAuthenticationProvider provider = new DaoAuthenticationProvider();
provider.setPasswordEncoder(passwordEncoder);
provider.setUserDetailsService(detailsService);
return provider;
}
}
@Configuration
@Order(2)
public class AdminWebSecurityConfiguration extends WebSecurityConfigurerAdapter {
@Resource
private AdminDetailsServiceImpl detailsService;
@Resource
private AdminLoginAuthorizationTokenFilter tokenFilter;
private final String[] authWhitelist = new String[]{"/admin/login", "/admin/auth/login"};
@Override
protected void configure(AuthenticationManagerBuilder auth) throws Exception {
auth
.userDetailsService(detailsService)
.passwordEncoder(passwordEncoder);
}
@Override
protected void configure(HttpSecurity http) throws Exception {
http
.requestMatcher(new AntPathRequestMatcher("/admin/**"))
.csrf(AbstractHttpConfigurer::disable)
.httpBasic(AbstractHttpConfigurer::disable)
.formLogin(n -> n
.usernameParameter("mail")
.loginPage("/admin/login")
.loginProcessingUrl("/admin/auth/login")
.successHandler(successHandler)
.failureHandler(failureHandler))
.logout(n -> n.logoutUrl("/admin/auth/logout"))
.exceptionHandling(n->n.authenticationEntryPoint(entryPoint))
.sessionManagement(n->n.sessionCreationPolicy(SessionCreationPolicy.STATELESS))
.addFilterBefore(tokenFilter, UsernamePasswordAuthenticationFilter.class)
.authorizeRequests(n -> n
.antMatchers(authWhitelist).permitAll()
.antMatchers("/admin/api/**").authenticated()
.anyRequest().denyAll());
}
/**
* 管理员的登录处理器
* @return 登录处理器
*/
public AuthenticationProvider adminAuthenticationProvider() {
DaoAuthenticationProvider provider = new DaoAuthenticationProvider();
provider.setPasswordEncoder(passwordEncoder);
provider.setUserDetailsService(detailsService);
return provider;
}
}
/**
* 注册密码加密器
*
* @return 密码加密器
*/
@Bean
public PasswordEncoder bCryptPasswordEncoder() {
return new BCryptPasswordEncoder();
}
}
在Security中并没有找到符合要求的请求匹配器，所以自己实现了一个SkipPathAntMatcher匹配器，过滤除指定请求路径外的所有请求，其实现方式复制参考的AntPathRequestMatcher，实现就是给AntPathRequestMatcher中的比较添加了个取反。
package com.nineya.user.config;
import java.util.Collections;
import java.util.Map;
import javax.servlet.http.HttpServletRequest;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.springframework.http.HttpMethod;
import org.springframework.security.web.util.matcher.RequestMatcher;
import org.springframework.security.web.util.matcher.RequestVariablesExtractor;
import org.springframework.util.AntPathMatcher;
import org.springframework.util.Assert;
import org.springframework.util.StringUtils;
import org.springframework.web.util.UrlPathHelper;
/**
* 排除选中的请求路径
*
* @author 殇雪话诀别
* 2020/11/28
*/
public final class SkipPathAntMatcher
implements RequestMatcher, RequestVariablesExtractor {
private static final Log logger = LogFactory.getLog(SkipPathAntMatcher.class);
private static final String MATCH_ALL = "/**";
private final Matcher matcher;
private final String pattern;
private final HttpMethod httpMethod;
private final boolean caseSensitive;
private final UrlPathHelper urlPathHelper;
/**
* 创建具有特定模式的匹配器，该模式将以区分大小写的方式排除所有HTTP路径。
*
* @param pattern 排除的路径
*/
public SkipPathAntMatcher(String pattern) {
this(pattern, null);
}
/**
* 以区分大小写的方式使用提供的模式和HTTP方法创建匹配器，只允许匹配的请求模式，不匹配的路径请求。
*
* @param pattern 排除的路径
* @param httpMethod 请求模式
*/
public SkipPathAntMatcher(String pattern, String httpMethod) {
this(pattern, httpMethod, true);
}
/**
* 以区分大小写的方式使用提供的模式和HTTP方法创建匹配器，指定路径是否判断大小写，只允许请求匹配的模式，不匹配的路径请求。
*
* @param pattern 排除的路径
* @param httpMethod 请求模式
* @param caseSensitive 是否考虑大小写
*/
public SkipPathAntMatcher(String pattern, String httpMethod,
boolean caseSensitive) {
this(pattern, httpMethod, caseSensitive, null);
}
/**
* 以区分大小写的方式使用提供的模式和HTTP方法创建匹配器，指定路径是否判断大小写，只允许请求匹配的模式，不匹配的路径请求。
*
* @param pattern 排除的路径
* @param httpMethod 请求模式
* @param caseSensitive 是否考虑大小写
* @param urlPathHelper 如果非空，将用于从HttpServletRequest提取路径
*/
public SkipPathAntMatcher(String pattern, String httpMethod,
boolean caseSensitive, UrlPathHelper urlPathHelper) {
Assert.hasText(pattern, "Pattern cannot be null or empty");
this.caseSensitive = caseSensitive;
if (pattern.equals(MATCH_ALL) || pattern.equals("**")) {
pattern = MATCH_ALL;
this.matcher = null;
} else {
// If the pattern ends with {@code /**} and has no other wildcards or path
// variables, then optimize to a sub-path match
if (pattern.endsWith(MATCH_ALL)
&& (pattern.indexOf('?') == -1 && pattern.indexOf('{') == -1
&& pattern.indexOf('}') == -1)
&& pattern.indexOf("*") == pattern.length() - 2) {
this.matcher = new SkipPathAntMatcher.SubpathMatcher(
pattern.substring(0, pattern.length() - 3), caseSensitive);
} else {
this.matcher = new SkipPathAntMatcher.SpringAntMatcher(pattern, caseSensitive);
}
}
this.pattern = pattern;
this.httpMethod = StringUtils.hasText(httpMethod) ? HttpMethod.valueOf(httpMethod)
: null;
this.urlPathHelper = urlPathHelper;
}
/**
* 如果配置的模式（和HTTP方法）与提供的请求的模式匹配，则返回false
*
* @param request 要匹配的请求。ant模式将与请求的{@code servletPath} + {@code pathInfo}进行匹配。
*/
@Override
public boolean matches(HttpServletRequest request) {
// 请求模式是否匹配
if (this.httpMethod != null && StringUtils.hasText(request.getMethod())
&& this.httpMethod != valueOf(request.getMethod())) {
if (logger.isDebugEnabled()) {
logger.debug("Request '" + request.getMethod() + " "
+ getRequestPath(request) + "'" + " doesn't match '"
+ this.httpMethod + " " + this.pattern + "'");
}
return false;
}
// 请求设置的是全局，全部拦截
if (this.pattern.equals(MATCH_ALL)) {
if (logger.isDebugEnabled()) {
logger.debug("Request '" + getRequestPath(request)
+ "' matched by universal pattern '/**'");
}
return false;
}
String url = getRequestPath(request);
if (logger.isDebugEnabled()) {
logger.debug("Checking match of request : '" + url + "'; against '"
+ this.pattern + "'");
}
System.out.println(url + " "+ !this.matcher.matches(url));
// 响应不匹配的链接
return !this.matcher.matches(url);
}
@Override
@Deprecated
public Map<String, String> extractUriTemplateVariables(HttpServletRequest request) {
return matcher(request).getVariables();
}
@Override
public RequestMatcher.MatchResult matcher(HttpServletRequest request) {
if (this.matcher == null || !matches(request)) {
return MatchResult.notMatch();
}
String url = getRequestPath(request);
return MatchResult.match(this.matcher.extractUriTemplateVariables(url));
}
private String getRequestPath(HttpServletRequest request) {
if (this.urlPathHelper != null) {
return this.urlPathHelper.getPathWithinApplication(request);
}
String url = request.getServletPath();
String pathInfo = request.getPathInfo();
if (pathInfo != null) {
url = StringUtils.hasLength(url) ? url + pathInfo : pathInfo;
}
return url;
}
public String getPattern() {
return this.pattern;
}
@Override
public boolean equals(Object obj) {
if (!(obj instanceof SkipPathAntMatcher)) {
return false;
}
SkipPathAntMatcher other = (SkipPathAntMatcher) obj;
return this.pattern.equals(other.pattern) && this.httpMethod == other.httpMethod
&& this.caseSensitive == other.caseSensitive;
}
@Override
public int hashCode() {
int result = this.pattern != null ? this.pattern.hashCode() : 0;
result = 31 * result + (this.httpMethod != null ? this.httpMethod.hashCode() : 0);
result = 31 * result + (this.caseSensitive ? 1231 : 1237);
return result;
}
@Override
public String toString() {
StringBuilder sb = new StringBuilder();
sb.append("Ant [pattern='").append(this.pattern).append("'");
if (this.httpMethod != null) {
sb.append(", ").append(this.httpMethod);
}
sb.append("]");
return sb.toString();
}
/**
* Provides a save way of obtaining the HttpMethod from a String. If the method is
* invalid, returns null.
*
* @param method the HTTP method to use.
* @return the HttpMethod or null if method is invalid.
*/
private static HttpMethod valueOf(String method) {
try {
return HttpMethod.valueOf(method);
} catch (IllegalArgumentException e) {
}
return null;
}
private interface Matcher {
boolean matches(String path);
Map<String, String> extractUriTemplateVariables(String path);
}
private static class SpringAntMatcher implements SkipPathAntMatcher.Matcher {
private final AntPathMatcher antMatcher;
private final String pattern;
private SpringAntMatcher(String pattern, boolean caseSensitive) {
this.pattern = pattern;
this.antMatcher = createMatcher(caseSensitive);
}
@Override
public boolean matches(String path) {
return this.antMatcher.match(this.pattern, path);
}
@Override
public Map<String, String> extractUriTemplateVariables(String path) {
return this.antMatcher.extractUriTemplateVariables(this.pattern, path);
}
private static AntPathMatcher createMatcher(boolean caseSensitive) {
AntPathMatcher matcher = new AntPathMatcher();
matcher.setTrimTokens(false);
matcher.setCaseSensitive(caseSensitive);
return matcher;
}
}
/**
* 针对尾随通配符的优化匹配器
*/
private static class SubpathMatcher implements SkipPathAntMatcher.Matcher {
private final String subpath;
private final int length;
private final boolean caseSensitive;
private SubpathMatcher(String subpath, boolean caseSensitive) {
assert !subpath.contains("*");
this.subpath = caseSensitive ? subpath : subpath.toLowerCase();
this.length = subpath.length();
this.caseSensitive = caseSensitive;
}
@Override
public boolean matches(String path) {
if (!this.caseSensitive) {
path = path.toLowerCase();
}
return path.startsWith(this.subpath)
&& (path.length() == this.length || path.charAt(this.length) == '/');
}
@Override
public Map<String, String> extractUriTemplateVariables(String path) {
return Collections.emptyMap();
}
}
}