SpringMVC完成初始化流程之后,就进入Servlet标准生命周期的第二个阶段,即“service”阶段。在“service”阶段中,每一次Http请求到来,容器都会启动一个请求线程,通过service()方法,委派到doGet()或者doPost()这些方法,完成Http请求的处理。
在初始化流程中,SpringMVC巧妙的运用依赖注入读取参数,并最终建立一个与容器上下文相关联的Spring子上下文。这个子上下文,就像Struts2中xwork容器一样,为接下来的Http处理流程中各种编程元素提供了容身之所。如果说将Spring上下文关联到Servlet容器中,是SpringMVC框架的第一个亮点,那么在请求转发流程中,SpringMVC对各种处理环节编程元素的抽象,就是另外一个独具匠心的亮点。
Struts2采取的是一种完全和Web容器隔离和解耦的事件机制。诸如Action对象、Result对象、Interceptor对象,这些都是完全脱离Servlet容器的编程元素。Struts2将数据流和事件处理完全剥离开来,从Http请求中读取数据后,下面的事件处理流程就只依赖于这些数据,而完全不知道有Web环境的存在。
反观SpringMVC,无论HandlerMapping对象、HandlerAdapter对象还是View对象,这些核心的接口所定义的方法中,HttpServletRequest和HttpServletResponse对象都是直接作为方法的参数出现的。这也就意味着,框架的设计者,直接将SpringMVC框架和容器绑定到了一起。或者说,整个SpringMVC框架,都是依托着Servlet容器元素来设计的。下面就来看一下,源码中是如何体现这一点的。
1.请求转发的入口
就像任何一个注册在容器中的Servlet一样,DispatcherServlet也是通过自己的service()方法来接收和转发Http请求到具体的doGet()或doPost()这些方法的。以一次典型的GET请求为例,经过HttpServlet基类中service()方法的委派,请求会被转发到doGet()方法中。doGet()方法,在DispatcherServlet的父类FrameworkServlet类中被覆写。
@Override protected final void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException { processRequest(request, response); }
protected final void processRequest(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException { long startTime = System.currentTimeMillis(); Throwable failureCause = null; // Expose current LocaleResolver and request as LocaleContext. LocaleContext previousLocaleContext = LocaleContextHolder.getLocaleContext(); LocaleContextHolder.setLocaleContext(buildLocaleContext(request), this.threadContextInheritable); // Expose current RequestAttributes to current thread. RequestAttributes previousRequestAttributes = RequestContextHolder.getRequestAttributes(); ServletRequestAttributes requestAttributes = null; if (previousRequestAttributes == null || previousRequestAttributes.getClass().equals(ServletRequestAttributes.class)) { requestAttributes = new ServletRequestAttributes(request); RequestContextHolder.setRequestAttributes(requestAttributes, this.threadContextInheritable); } if (logger.isTraceEnabled()) { logger.trace("Bound request context to thread: " + request); } try { doService(request, response); } catch (ServletException ex) { failureCause = ex; throw ex; } catch (IOException ex) { failureCause = ex; throw ex; } catch (Throwable ex) { failureCause = ex; throw new NestedServletException("Request processing failed", ex); } finally { // Clear request attributes and reset thread-bound context. LocaleContextHolder.setLocaleContext(previousLocaleContext, this.threadContextInheritable); if (requestAttributes != null) { RequestContextHolder.setRequestAttributes(previousRequestAttributes, this.threadContextInheritable); requestAttributes.requestCompleted(); } if (logger.isTraceEnabled()) { logger.trace("Cleared thread-bound request context: " + request); } if (logger.isDebugEnabled()) { if (failureCause != null) { this.logger.debug("Could not complete request", failureCause); } else { this.logger.debug("Successfully completed request"); } } if (this.publishEvents) { // Whether or not we succeeded, publish an event. long processingTime = System.currentTimeMillis() - startTime; this.webApplicationContext.publishEvent( new ServletRequestHandledEvent(this, request.getRequestURI(), request.getRemoteAddr(), request.getMethod(), getServletConfig().getServletName(), WebUtils.getSessionId(request), getUsernameForRequest(request), processingTime, failureCause)); } } }
可以看到,processRequest()方法只是做了一些线程安全的隔离,真正的请求处理,发生在doService()方法中。点开FrameworkServlet类中的doService()方法。
protected abstract void doService(HttpServletRequest request, HttpServletResponse response) throws Exception;
让我们接着看DispatcherServlet类中实现的doService()方法。
@Override protected void doService(HttpServletRequest request, HttpServletResponse response) throws Exception { if (logger.isDebugEnabled()) { String requestUri = urlPathHelper.getRequestUri(request); logger.debug("DispatcherServlet with name '" + getServletName() + "' processing " + request.getMethod() + " request for [" + requestUri + "]"); } // Keep a snapshot of the request attributes in case of an include, // to be able to restore the original attributes after the include. Map<String, Object> attributesSnapshot = null; if (WebUtils.isIncludeRequest(request)) { logger.debug("Taking snapshot of request attributes before include"); attributesSnapshot = new HashMap<String, Object>(); Enumeration<?> attrNames = request.getAttributeNames(); while (attrNames.hasMoreElements()) { String attrName = (String) attrNames.nextElement(); if (this.cleanupAfterInclude || attrName.startsWith("org.springframework.web.servlet")) { attributesSnapshot.put(attrName, request.getAttribute(attrName)); } } } // Make framework objects available to handlers and view objects. request.setAttribute(WEB_APPLICATION_CONTEXT_ATTRIBUTE, getWebApplicationContext()); request.setAttribute(LOCALE_RESOLVER_ATTRIBUTE, this.localeResolver); request.setAttribute(THEME_RESOLVER_ATTRIBUTE, this.themeResolver); request.setAttribute(THEME_SOURCE_ATTRIBUTE, getThemeSource()); FlashMap inputFlashMap = this.flashMapManager.retrieveAndUpdate(request, response); if (inputFlashMap != null) { request.setAttribute(INPUT_FLASH_MAP_ATTRIBUTE, Collections.unmodifiableMap(inputFlashMap)); } request.setAttribute(OUTPUT_FLASH_MAP_ATTRIBUTE, new FlashMap()); request.setAttribute(FLASH_MAP_MANAGER_ATTRIBUTE, this.flashMapManager); try { doDispatch(request, response); } finally { // Restore the original attribute snapshot, in case of an include. if (attributesSnapshot != null) { restoreAttributesAfterInclude(request, attributesSnapshot); } } }
所以,你可以看到,真正发生请求转发的方法doDispatch()中,它的参数是HttpServletRequest和HttpServletResponse对象。这给我们传递的意思也很明确,从request中能获取到一切请求的数据,从response中,我们又可以往服务器端输出任何响应,Http请求的处理,就应该围绕这两个对象来设计。我们不妨可以将SpringMVC这种设计方案,是从Struts2的过度设计中吸取教训,而向Servlet编程的一种回归和简化。
2.请求转发的抽象描述
接下来让我们看看doDispatch()这个整个请求转发流程中最核心的方法。DispatcherServlet所接收的Http请求,经过层层转发,最终都是汇总到这个方法中来进行最后的请求分发和处理。doDispatch()这个方法的内容,就是SpringMVC整个框架的精华所在。它通过高度抽象的接口,描述出了一个MVC(Model-View-Controller)设计模式的实现方案。Model、View、Controller三种层次的编程元素,在SpringMVC中都有大量的实现类,各种处理细节也是千差万别。但是,它们最后都是由,也都能由doDispatch()方法来统一描述,这就是接口和抽象的威力,万变不离其宗。
先来看一下doDispatch()方法的庐山真面目。
protected void doDispatch(HttpServletRequest request, HttpServletResponse response) throws Exception { HttpServletRequest processedRequest = request; HandlerExecutionChain mappedHandler = null; int interceptorIndex = -1; try { ModelAndView mv; boolean errorView = false; try { processedRequest = checkMultipart(request); // Determine handler for the current request. mappedHandler = getHandler(processedRequest, false); if (mappedHandler == null || mappedHandler.getHandler() == null) { noHandlerFound(processedRequest, response); return; } // Determine handler adapter for the current request. HandlerAdapter ha = getHandlerAdapter(mappedHandler.getHandler()); // Process last-modified header, if supported by the handler. String method = request.getMethod(); boolean isGet = "GET".equals(method); if (isGet || "HEAD".equals(method)) { long lastModified = ha.getLastModified(request, mappedHandler.getHandler()); if (logger.isDebugEnabled()) { String requestUri = urlPathHelper.getRequestUri(request); logger.debug("Last-Modified value for [" + requestUri + "] is: " + lastModified); } if (new ServletWebRequest(request, response).checkNotModified(lastModified) && isGet) { return; } } // Apply preHandle methods of registered interceptors. HandlerInterceptor[] interceptors = mappedHandler.getInterceptors(); if (interceptors != null) { for (int i = 0; i < interceptors.length; i++) { HandlerInterceptor interceptor = interceptors[i]; if (!interceptor.preHandle(processedRequest, response, mappedHandler.getHandler())) { triggerAfterCompletion(mappedHandler, interceptorIndex, processedRequest, response, null); return; } interceptorIndex = i; } } // Actually invoke the handler. mv = ha.handle(processedRequest, response, mappedHandler.getHandler()); // Do we need view name translation? if (mv != null && !mv.hasView()) { mv.setViewName(getDefaultViewName(request)); } // Apply postHandle methods of registered interceptors. if (interceptors != null) { for (int i = interceptors.length - 1; i >= 0; i--) { HandlerInterceptor interceptor = interceptors[i]; interceptor.postHandle(processedRequest, response, mappedHandler.getHandler(), mv); } } } catch (ModelAndViewDefiningException ex) { logger.debug("ModelAndViewDefiningException encountered", ex); mv = ex.getModelAndView(); } catch (Exception ex) { Object handler = (mappedHandler != null ? mappedHandler.getHandler() : null); mv = processHandlerException(processedRequest, response, handler, ex); errorView = (mv != null); } // Did the handler return a view to render? if (mv != null && !mv.wasCleared()) { render(mv, processedRequest, response); if (errorView) { WebUtils.clearErrorRequestAttributes(request); } } else { if (logger.isDebugEnabled()) { logger.debug("Null ModelAndView returned to DispatcherServlet with name '" + getServletName() + "': assuming HandlerAdapter completed request handling"); } } // Trigger after-completion for successful outcome. triggerAfterCompletion(mappedHandler, interceptorIndex, processedRequest, response, null); } catch (Exception ex) { // Trigger after-completion for thrown exception. triggerAfterCompletion(mappedHandler, interceptorIndex, processedRequest, response, ex); throw ex; } catch (Error err) { ServletException ex = new NestedServletException("Handler processing failed", err); // Trigger after-completion for thrown exception. triggerAfterCompletion(mappedHandler, interceptorIndex, processedRequest, response, ex); throw ex; } finally { // Clean up any resources used by a multipart request. if (processedRequest != request) { cleanupMultipart(processedRequest); } } }真是千呼万唤始出来,犹抱琵琶半遮面。我们在第一篇 《SpringMVC源码剖析(一)- 从抽象和接口说起》中所描述的各种编程元素,依次出现在该方法中。HandlerMapping、HandlerAdapter、View这些接口的设计,我们在第一篇中已经讲过。现在我们来重点关注一下HandlerExecutionChain这个对象。
从上面的代码中,很明显可以看出一条线索,整个方法是围绕着如何获取HandlerExecutionChain对象,执行HandlerExecutionChain对象得到相应的视图对象,再对视图进行渲染这条主线来展开的。HandlerExecutionChain对象显得异常重要。
因为Http请求要进入SpringMVC的处理体系,必须由HandlerMapping接口的实现类映射Http请求,得到一个封装后的HandlerExecutionChain对象。再由HandlerAdapter接口的实现类来处理这个HandlerExecutionChain对象所包装的处理对象,来得到最后渲染的视图对象。
视图对象是用ModelAndView对象来描述的,名字已经非常直白,就是数据和视图,其中的数据,由HttpServletRequest的属性得到,视图就是由HandlerExecutionChain封装的处理对象处理后得到。当然HandlerExecutionChain中的拦截器列表HandlerInterceptor,会在处理过程的前后依次被调用,为处理过程留下充足的扩展点。
所有的SpringMVC框架元素,都是围绕着HandlerExecutionChain这个执行链来发挥效用。我们来看看,HandlerExecutionChain类的代码。
package org.springframework.web.servlet; import java.util.ArrayList; import java.util.Arrays; import java.util.List; import org.springframework.util.CollectionUtils; public class HandlerExecutionChain { private final Object handler; private HandlerInterceptor[] interceptors; private List<HandlerInterceptor> interceptorList; public HandlerExecutionChain(Object handler) { this(handler, null); } public HandlerExecutionChain(Object handler, HandlerInterceptor[] interceptors) { if (handler instanceof HandlerExecutionChain) { HandlerExecutionChain originalChain = (HandlerExecutionChain) handler; this.handler = originalChain.getHandler(); this.interceptorList = new ArrayList<HandlerInterceptor>(); CollectionUtils.mergeArrayIntoCollection(originalChain.getInterceptors(), this.interceptorList); CollectionUtils.mergeArrayIntoCollection(interceptors, this.interceptorList); } else { this.handler = handler; this.interceptors = interceptors; } } public Object getHandler() { return this.handler; } public void addInterceptor(HandlerInterceptor interceptor) { initInterceptorList(); this.interceptorList.add(interceptor); } public void addInterceptors(HandlerInterceptor[] interceptors) { if (interceptors != null) { initInterceptorList(); this.interceptorList.addAll(Arrays.asList(interceptors)); } } private void initInterceptorList() { if (this.interceptorList == null) { this.interceptorList = new ArrayList<HandlerInterceptor>(); } if (this.interceptors != null) { this.interceptorList.addAll(Arrays.asList(this.interceptors)); this.interceptors = null; } } public HandlerInterceptor[] getInterceptors() { if (this.interceptors == null && this.interceptorList != null) { this.interceptors = this.interceptorList.toArray(new HandlerInterceptor[this.interceptorList.size()]); } return this.interceptors; } @Override public String toString() { if (this.handler == null) { return "HandlerExecutionChain with no handler"; } StringBuilder sb = new StringBuilder(); sb.append("HandlerExecutionChain with handler [").append(this.handler).append("]"); if (!CollectionUtils.isEmpty(this.interceptorList)) { sb.append(" and ").append(this.interceptorList.size()).append(" interceptor"); if (this.interceptorList.size() > 1) { sb.append("s"); } } return sb.toString(); } }
1.拦截器组成的列表,在执行对象被调用的前后,会依次执行。这里可以看成是一个的AOP环绕通知,拦截器可以对处理对象随心所欲的进行处理和增强。这里明显是吸收了Struts2中拦截器的设计思想。这种AOP环绕式的扩展点设计,也几乎成为所有框架必备的内容。
2.实际的处理对象,即handler对象,是由Object对象来引用的。
<span style="font-size:14px;">private final Object handler;</span>
我们常说,一个框架最高层次的抽象是接口,但是这里SpringMVC更进了一步。在最后的处理对象上面,SpringMVC没有对它做任何的限制,只要是java世界中的对象,都可以用来作为最后的处理对象,来生成视图。极端一点来说,你甚至可以将另外一个MVC框架集成到SpringMVC中来,也就是为什么SpringMVC官方文档中,居然还有集成其他表现层框架的内容。这一点,在所有表现层框架中,是独领风骚,冠绝群雄的。
3.结语
SpringMVC的成功,源于它对开闭原则的运用和遵守。也正因此,才使得整个框架具有如此强大的描述和扩展能力。这也许和SpringMVC出现和兴起的时间有关,正是经历了Struts1到Struts2这些Web开发领域MVC框架的更新换代,它的设计者才能站在前人的肩膀上。知道了如何将事情做的糟糕之后,你或许才知道如何将事情做得好。
希望在这个系列里面分享的SpringMVC源码阅读经验,能帮助读者们从更高的层次来审视SpringMVC框架的设计,也希望这里所描述的一些基本设计思想,能在你更深入的了解SpringMVC的细节时,对你有帮助。哲学才是唯一的、最终的武器,在一个框架的设计上,尤其是如此。经常地体会一个框架设计者的设计思想,对你更好的使用它,是有莫大的益处的。
本文转载自:SpringMVC源码剖析(四)- DispatcherServlet请求转发的实现