前言
在使用Spring和SpringMVC的老版本进行开发时,我们需要配置很多的xml文件,非常的繁琐,总是让用户自行选择配置也是非常不好的。基于约定大于配置的规定,Spring提供了很多注解帮助我们简化了大量的xml配置;但是在使用SpringMVC时,我们还会使用到WEB-INF/web.xml,但实际上我们是完全可以使用Java类来取代xml配置的,这也是后来SpringBoott的实现原理。本篇就来看看Spring是如何实现完全的零XML配置。
正文
先来看一下原始的web.xml配置:
contextConfigLocation classpath:spring.xml webAppRootKey ServicePlatform.root org.springframework.web.context.ContextLoaderListener spring-dispatcher org.springframework.web.servlet.DispatcherServlet contextConfigLocation classpath:spring-dispatcher.xml 0 spring-dispatcher /
这里各个配置的作用简单说下,context-param是加载我们主的sping.xml配置,比如一些bean的配置和开启注解扫描等;listener是配置监听器,Tomcat启动会触发监听器调用;servlet则是配置我们自定义的Servlet实现,比如DispatcherServlet。还有其它很多配置就不一一说明了,在这里主要看到记住context-param和servlet配置,这是SpringIOC父子容器的体现。
在之前的I文章中讲过IOC容器是以父子关系组织的,但估计大部分人都不能理解,除了看到复杂的继承体系,并没有看到父容器作用的体现,稍后来分析。
了解了配置,我们就需要思考如何替换掉这些繁琐的配置。实际上Tomcat提供了一个规范,有一个ServletContainerInitializer接口:
public interface ServletContainerInitializer { void onStartup(Set> var1, ServletContext var2) throws ServletException; }
Tomcat启动时会调用该接口实现类的onStartup方法,这个方法有两个参数,第二个不用说,主要是第一个参数什么?从哪里来?另外我们自定义的实现类又怎么让Tomcat调用呢?
首先解答最后一个问题,这里也是利用SPI来实现的,因此我们实现了该接口后,还需要在META-INF.services下配置。其次,这里传入的第一个参数也是我们自定义的扩展接口的实现类,我们可以通过我们自定义的接口实现很多需要在启动时做的事,比如加载Servlet,但是Tomcat又是怎么知道我们自定义的接口是哪个呢?
这就需要用到@HandlesTypes注解,该注解就是标注在ServletContainerInitializer的实现类上,其值就是我们扩展的接口,这样Tomcat就知道需要传入哪个接口实现类到这个onStartup方法了。
来看一个简单的实现:
@HandlesTypes(LoadServlet.class) public class MyServletContainerInitializer implements ServletContainerInitializer { @Override public void onStartup(Set> set, ServletContext servletContext) throws ServletException { Iterator var4; if (set != null) { var4 = set.iterator(); while (var4.hasNext()) { Class> clazz = (Class>) var4.next(); if (!clazz.isInterface() && !Modifier.isAbstract(clazz.getModifiers()) && LoadServlet.class.isAssignableFrom(clazz)) { try { ((LoadServlet) clazz.newInstance()).loadOnstarp(servletContext); } catch (Exception e) { e.printStackTrace(); } } } } } } public interface LoadServlet { void loadOnstarp(ServletContext servletContext); } public class LoadServletImpl implements LoadServlet { @Override public void loadOnstarp(ServletContext servletContext) { ServletRegistration.Dynamic initServlet = servletContext.addServlet("initServlet", "org.springframework.web.servlet.DispatcherServlet"); initServlet.setLoadOnStartup(1); initServlet.addMapping("/init"); } }
这就是Tomcat给我们提供的规范,通过这个规范我们就能实现Spring的零xml配置启动,直接来看Spring是如何做的。根据上面所说我们可以在spring-web工程下找到META-INF/services/javax.servlet.ServletContainerInitializer配置:
@HandlesTypes(WebApplicationInitializer.class) public class SpringServletContainerInitializer implements ServletContainerInitializer { @Override public void onStartup(@Nullable Set> webAppInitializerClasses, ServletContext servletContext) throws ServletException { List initializers = new LinkedList<>(); if (webAppInitializerClasses != null) { for (Class> waiClass : webAppInitializerClasses) { // Be defensive: Some servlet containers provide us with invalid classes, // no matter what @HandlesTypes says... if (!waiClass.isInterface() && !Modifier.isAbstract(waiClass.getModifiers()) && WebApplicationInitializer.class.isAssignableFrom(waiClass)) { try { initializers.add((WebApplicationInitializer) ReflectionUtils.accessibleConstructor(waiClass).newInstance()); } catch (Throwable ex) { throw new ServletException("Failed to instantiate WebApplicationInitializer class", ex); } } } } if (initializers.isEmpty()) { servletContext.log("No Spring WebApplicationInitializer types detected on classpath"); return; } servletContext.log(initializers.size() + " Spring WebApplicationInitializers detected on classpath"); AnnotationAwareOrderComparator.sort(initializers); for (WebApplicationInitializer initializer : initializers) { initializer.onStartup(servletContext); } } }
核心的实现就是WebApplicationInitializer,先看看其继承体系
AbstractReactiveWebInitializer不用管,主要看另外一边,但是都是抽象类,也就是说真的实例也是由我们自己实现,但需要我们实现什么呢?我们一般直接继承AbstractAnnotationConfigDispatcherServletInitializer类,有四个抽象方法需要我们实现:
//父容器 @Override protected Class>[] getRootConfigClasses() { return new Class>[]{SpringContainer.class}; } //SpringMVC配置子容器 @Override protected Class>[] getServletConfigClasses() { return new Class>[]{MvcContainer.class}; } //获取DispatcherServlet的映射信息 @Override protected String[] getServletMappings() { return new String[]{"/"}; } // filter配置 @Override protected Filter[] getServletFilters() { MyFilter myFilter = new MyFilter(); CorsFilter corsFilter = new CorsFilter(); return new Filter[]{myFilter,corsFilter}; }
这里主要注意getRootConfigClasses和getServletConfigClasses方法,分别加载父、子容器:
@ComponentScan(value = "com.dark",excludeFilters = { @ComponentScan.Filter(type = FilterType.ANNOTATION,classes = {Controller.class}) }) public class SpringContainer { } @ComponentScan(value = "com.dark",includeFilters = { @ComponentScan.Filter(type = FilterType.ANNOTATION,classes = {Controller.class}) },useDefaultFilters = false) public class MvcContainer { }
看到这两个类上的注解应该不陌生了吧,父容器扫描装载了所有不带@Controller注解的类,子容器则相反,但需要对象时首先从当前容器中找,如果没有则从父容器中获取,为什么要这么设计呢?
直接放到一个容器中不行么?先思考下, 稍后解答。回到onStartup方法中,直接回调用到AbstractDispatcherServletInitializer类:
public void onStartup(ServletContext servletContext) throws ServletException { super.onStartup(servletContext); //注册DispatcherServlet registerDispatcherServlet(servletContext); }
先是调用父类:
public void onStartup(ServletContext servletContext) throws ServletException { registerContextLoaderListener(servletContext); } protected void registerContextLoaderListener(ServletContext servletContext) { //创建spring上下文,注册了SpringContainer WebApplicationContext rootAppContext = createRootApplicationContext(); if (rootAppContext != null) { //创建监听器 ContextLoaderListener listener = new ContextLoaderListener(rootAppContext); listener.setContextInitializers(getRootApplicationContextInitializers()); servletContext.addListener(listener); } }
然后调用createRootApplicationContext创建父容器:
protected WebApplicationContext createRootApplicationContext() { Class>[] configClasses = getRootConfigClasses(); if (!ObjectUtils.isEmpty(configClasses)) { AnnotationConfigWebApplicationContext context = new AnnotationConfigWebApplicationContext(); context.register(configClasses); return context; } else { return null; } }
可以看到就是创建了一个AnnotationConfigWebApplicationContext对象,并将我们的配置类SpringContainer注册了进去。接着创建Tomcat启动加载监听器ContextLoaderListener,该监听器有一个contextInitialized方法,会在Tomcat启动时调用。
public void contextInitialized(ServletContextEvent event) { initWebApplicationContext(event.getServletContext()); } */ public WebApplicationContext initWebApplicationContext(ServletContext servletContext) { long startTime = System.currentTimeMillis(); try { // Store context in local instance variable, to guarantee that // it is available on ServletContext shutdown. if (this.context == null) { this.context = createWebApplicationContext(servletContext); } if (this.context instanceof ConfigurableWebApplicationContext) { ConfigurableWebApplicationContext cwac = (ConfigurableWebApplicationContext) this.context; if (!cwac.isActive()) { // The context has not yet been refreshed -> provide services such as // setting the parent context, setting the application context id, etc if (cwac.getParent() == null) { // The context instance was injected without an explicit parent -> // determine parent for root web application context, if any. ApplicationContext parent = loadParentContext(servletContext); cwac.setParent(parent); } configureAndRefreshWebApplicationContext(cwac, servletContext); } } servletContext.setAttribute(WebApplicationContext.ROOT_WEB_APPLICATION_CONTEXT_ATTRIBUTE, this.context); ClassLoader ccl = Thread.currentThread().getContextClassLoader(); if (ccl == ContextLoader.class.getClassLoader()) { currentContext = this.context; } else if (ccl != null) { currentContextPerThread.put(ccl, this.context); } return this.context; } }
可以看到就是去初始化容器,这个和之前分析xml解析是一样的,主要注意这里封装了ServletContext对象,并将父容器设置到了该对象中。
父容器创建完成后自然就是子容器的创建,来到registerDispatcherServlet方法:
protected void registerDispatcherServlet(ServletContext servletContext) { String servletName = getServletName(); Assert.hasLength(servletName, "getServletName() must not return null or empty"); //创建springmvc的上下文,注册了MvcContainer类 WebApplicationContext servletAppContext = createServletApplicationContext(); Assert.notNull(servletAppContext, "createServletApplicationContext() must not return null"); //创建DispatcherServlet FrameworkServlet dispatcherServlet = createDispatcherServlet(servletAppContext); Assert.notNull(dispatcherServlet, "createDispatcherServlet(WebApplicationContext) must not return null"); dispatcherServlet.setContextInitializers(getServletApplicationContextInitializers()); ServletRegistration.Dynamic registration = servletContext.addServlet(servletName, dispatcherServlet); if (registration == null) { throw new IllegalStateException("Failed to register servlet with name '" + servletName + "'. " + "Check if there is another servlet registered under the same name."); } /* * 如果该元素的值为负数或者没有设置,则容器会当Servlet被请求时再加载。 如果值为正整数或者0时,表示容器在应用启动时就加载并初始化这个servlet, 值越小,servlet的优先级越高,就越先被加载 * */ registration.setLoadOnStartup(1); registration.addMapping(getServletMappings()); registration.setAsyncSupported(isAsyncSupported()); Filter[] filters = getServletFilters(); if (!ObjectUtils.isEmpty(filters)) { for (Filter filter : filters) { registerServletFilter(servletContext, filter); } } customizeRegistration(registration); } protected WebApplicationContext createServletApplicationContext() { AnnotationConfigWebApplicationContext context = new AnnotationConfigWebApplicationContext(); Class>[] configClasses = getServletConfigClasses(); if (!ObjectUtils.isEmpty(configClasses)) { context.register(configClasses); } return context; }
这里也是创建了一个AnnotationConfigWebApplicationContext对象,不同的只是这里注册的配置类就是我们的Servlet配置了。然后创建了DispatcherServlet对象,并将上下文对象设置了进去。
看到这你可能会疑惑,既然父子容器创建的都是相同类的对象,何来的父子容器之说?
别急,这个在初始化该上文时就明白了。但是这里的初始化入口在哪呢?没有看到任何监听器的创建和调用。
实际上这里的上下文对象初始化是在Servlet初始化时实现的,即init方法,直接来到HttpServletBean的init方法(分析SpringMVC源码时讲过):
public final void init() throws ServletException { ...省略 // Let subclasses do whatever initialization they like. initServletBean(); } protected final void initServletBean() throws ServletException { try { this.webApplicationContext = initWebApplicationContext(); initFrameworkServlet(); } } protected WebApplicationContext initWebApplicationContext() { //这里会从servletContext中获取到父容器,就是通过监听器加载的容器 WebApplicationContext rootContext = WebApplicationContextUtils.getWebApplicationContext(getServletContext()); WebApplicationContext wac = null; if (this.webApplicationContext != null) { // A context instance was injected at construction time -> use it wac = this.webApplicationContext; if (wac instanceof ConfigurableWebApplicationContext) { ConfigurableWebApplicationContext cwac = (ConfigurableWebApplicationContext) wac; if (!cwac.isActive()) { if (cwac.getParent() == null) { cwac.setParent(rootContext); } //容器加载 configureAndRefreshWebApplicationContext(cwac); } } } if (wac == null) { wac = findWebApplicationContext(); } if (wac == null) { wac = createWebApplicationContext(rootContext); } if (!this.refreshEventReceived) { synchronized (this.onRefreshMonitor) { onRefresh(wac); } } if (this.publishContext) { // Publish the context as a servlet context attribute. String attrName = getServletContextAttributeName(); getServletContext().setAttribute(attrName, wac); } return wac; }
看到这里想你也应该明白了,首先从ServletContext中拿到父容器,然后设置到当前容器的parent中,实现了父子容器的组织,而这样设计好处我想也是很清楚的,子容器目前装载的都是MVC的配置和Bean,简单点说就是Controller,父容器中都是Service,Controller是依赖于Service的,如果不构建这样的层级关系并优先实例化父容器,你怎么实现Controller层的依赖注入成功呢?
总结
本篇结合之前的文章,分析了SpringMVC零XML配置的实现原理,也补充了之前未分析到父子容器关系,让我们能从细节上更加全面的理解SpringIOC的实现原理,相信看完本篇对于SpringBoot的实现你也会有自己的想法。希望能给大家一个参考,也希望大家多多支持脚本之家。