背景

本文作为 SpringMVC系列 第二篇，介绍HTTP请求的调用链：从请求进入Tomcat到数据流返回客户端的完整过程。为了尽可能把流程表达清楚，进行了很多减支处理，只关注主线逻辑。

本文也作为SpringMVC系列后续文章的基础，在调用链梳理清楚的基础上，后文对重要逻辑分别进行展开介绍，如拦截器、异常处理器、转换器、消息转换器、异步请求、文件上传等。在这些文章完成后，会出一个Spring框架应用专题，包括：结果集框架、错误码框架、鉴权逻辑、分页查询、事件框架等，基于此会对Spring系列和SpringMVC系列文章有更深层次的理解。

1.调用链

Tomcat从逻辑上可以分为连接器(coyote)和Servlet容器(catalina)两个部分：coyote负责接收客户端的请求，并按照协议对请求进行解析，封装成Java对象后发送给catalina以及将catalina返回的消息推送给客户端；catalina提供了Servlet容器实现，负责处理具体的请求并进行响应。
其中，coyote封装了底层的网络通讯(Socket)，为catalina提供了统一的接口(Request/Response对象)而与Servlet容器解耦；catalina内部通过适配器将(Request/Response对象)转换为(HttpRequest/HttpResponse对象)，然后将消息发送给Servlet对象，流程图如下所示：

总之，当Http请求到达Tomcat连接池后，会将请求消息封装成(HttpRequest/HttpResponse对象), 通过调用Servlet标准接口实现消息的传递。
SpringMVC框架对应的Servlet对象为DispatcherServlet，即调用栈会进入DispatcherServlet的void service(ServletRequest req, ServletResponse res)方法。
因此，有必要了解一下DispatcherServlet类的继承关系以及对Servlet方法实现情况，如下图所示：

2.HttpServlet

DispatcherServlet关于Servlet:void service(ServletRequest req, ServletResponse res)接口的实现逻辑在HttpServlet类中，代码如下：

public void service(ServletRequest req, ServletResponse res) throws ServletException, IOException {HttpServletRequest  request;HttpServletResponse response;try {request = (HttpServletRequest) req;response = (HttpServletResponse) res;} catch (ClassCastException e) {throw new ServletException(lStrings.getString("http.non_http"));}service(request, response);}

逻辑较为简单，直接将ServletRequest/ServletResponse对象转为HttpServletRequest/HttpServletResponse，并调用service(HttpServletRequest req, HttpServletResponse resp)接口。后者中根据HTTP方法类型派发给了doGet/doPost/doPut等接口；而doGet/doPost/doPut等接口的实现逻辑在FrameworkServlet中归一到void processRequest(HttpServletRequest request, HttpServletResponse response)接口中：

@Override
protected final void doGet(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {processRequest(request, response);
}@Override
protected final void doPost(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {processRequest(request, response);
}@Override
protected final void doPut(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {processRequest(request, response);
}

总之，所有来自Servlet的请求都会进入processRequest方法中进行处理。

3.FrameworkServlet

processRequest方法的主线逻辑如下：

protected final void processRequest(HttpServletRequest request, HttpServletResponse response) throws ServletException, IOException {LocaleContext previousLocaleContext = LocaleContextHolder.getLocaleContext();LocaleContext localeContext = buildLocaleContext(request);RequestAttributes previousAttributes = RequestContextHolder.getRequestAttributes();ServletRequestAttributes requestAttributes = buildRequestAttributes(request, response, previousAttributes);WebAsyncManager asyncManager = WebAsyncUtils.getAsyncManager(request);asyncManager.registerCallableInterceptor(FrameworkServlet.class.getName(), new RequestBindingInterceptor());initContextHolders(request, localeContext, requestAttributes);try {doService(request, response);} finally {resetContextHolders(request, previousLocaleContext, previousAttributes);}
}

上述代码在逻辑上可以分为三块：调用doService(request, response)接口处理并响应、调用前的准备工作、调用后的清理工作。
doService接口作为主体逻辑在下一节中进行介绍，本节对准备工作和清理工作进行介绍。

准备工作

LocaleContext previousLocaleContext = LocaleContextHolder.getLocaleContext();
LocaleContext localeContext = buildLocaleContext(request);RequestAttributes previousAttributes = RequestContextHolder.getRequestAttributes();
ServletRequestAttributes requestAttributes = buildRequestAttributes(request, response, previousAttributes);// WebAsyncManager提供了HTTP请求异步处理能力，不是本文的重点
WebAsyncManager asyncManager = WebAsyncUtils.getAsyncManager(request);
asyncManager.registerCallableInterceptor(FrameworkServlet.class.getName(), new RequestBindingInterceptor());

根据request对象获取Locale对象并将其通过ThreadLocal对象中，将request和response对象封装在RequestAttributes对象中，保存在ThreadLocal对象中；借助ThreadLocal的能力，对外提供了静态方法，使得程序在任意时刻都取得Locale对象、request和response对象。

清理工作

清理工作对应从内存中将该线程相关的Locale对象和request和response对象从内存中清除，以防止内存泄露。

resetContextHolders(request, previousLocaleContext, previousAttributes);private void resetContextHolders(HttpServletRequest request, LocaleContext prevLocaleContext, RequestAttributes previousAttributes) {LocaleContextHolder.setLocaleContext(prevLocaleContext, this.threadContextInheritable);RequestContextHolder.setRequestAttributes(previousAttributes, this.threadContextInheritable);
}

这是一个使用ThreadLocal基本的套路，后面介绍Spring应用专题时会涉及。

4.DispatchServlet

protected void doService(HttpServletRequest request, HttpServletResponse response) throws Exception {// springMVC容器request.setAttribute(WEB_APPLICATION_CONTEXT_ATTRIBUTE, getWebApplicationContext());// locale解析器request.setAttribute(LOCALE_RESOLVER_ATTRIBUTE, this.localeResolver);// 主体相关request.setAttribute(THEME_RESOLVER_ATTRIBUTE, this.themeResolver);request.setAttribute(THEME_SOURCE_ATTRIBUTE, getThemeSource());// flashMapManager在后续介绍转发和重定向内容时进行介绍，这里暂时忽略if (this.flashMapManager != null) {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 {// 清理工作 
}

doService逻辑比较简单，包括：调用doDispatch方法前对request的属性设置，以及在调用之后清理工作。doDispatch方法是核心，其他逻辑是在为调用该方法做的准备或收尾工作。
本文希望将HTTP调用链的主线逻辑表达清楚，因此会省去异步请求(WebAsyncManager相关)、文件上传(MultipartResolver相关)、异常捕获等逻辑，如下所示：

protected void doDispatch(HttpServletRequest request, HttpServletResponse response) throws Exception {HandlerExecutionChain mappedHandler = null;boolean multipartRequestParsed = false;try {ModelAndView mv = null;Exception dispatchException = null;try {mappedHandler = getHandler(request);if (mappedHandler == null) {noHandlerFound(request, response);return;}HandlerAdapter ha = getHandlerAdapter(mappedHandler.getHandler());if (!mappedHandler.applyPreHandle(request, response)) {return;}mv = ha.handle(request, response, mappedHandler.getHandler());mappedHandler.applyPostHandle(request, response, mv);} catch (Exception ex) {dispatchException = ex;}processDispatchResult(request, response, mappedHandler, mv, dispatchException);} catch (Exception ex) {triggerAfterCompletion(request, response, mappedHandler, ex);} 
}

上述逻辑可以表示为：

拦截器的preHandle方法按照拦截器在列表中的顺序正向执行，postHandle和afterCompletion反向执行。

上图可以分为以下几个步骤：
（1）根据request对象获取执行链HandlerExecutionChain；执行链由Interceptor(拦截器)和Handler(对controller的包装)组成;
（2）根据Handler获取HandlerAdapter；
（3）依次调用拦截器的preHandle方法；
（4）借助HandlerAdapter 通过反射调用目标方法(controller对应的方法)；
（5）依次调用拦截器的postHandle方法；
（6）对结果或者异常进行后置处理；
（7）依次调用拦截器的afterCompletion方法；

上述步骤为正常执行流程，异常场景包括：

异常场景1:
当根据equest对象获取执行链HandlerExecutionChain失败时(没有匹配项)，则直接想客户端返回404。

异常场景2:
调用某个拦截器的preHandle方法返回false时，则反向执行已执行过拦截器的afterCompletion方法，并跳出doDispatch方法。

涉及的代码逻辑如下：

protected void doDispatch(HttpServletRequest request, HttpServletResponse response) throws Exception {//...if (!mappedHandler.applyPreHandle(processedRequest, response)) {return;}//...
}boolean applyPreHandle(HttpServletRequest request, HttpServletResponse response) throws Exception {for (int i = 0; i < this.interceptorList.size(); i++) {HandlerInterceptor interceptor = this.interceptorList.get(i);if (!interceptor.preHandle(request, response, this.handler)) {triggerAfterCompletion(request, response, null);return false;}this.interceptorIndex = i;}return true;
}

当有拦截器的preHandle方法返回false时，调用triggerAfterCompletion(request, response, null)方法：

void triggerAfterCompletion(HttpServletRequest request, HttpServletResponse response, @Nullable Exception ex) {for (int i = this.interceptorIndex; i >= 0; i--) {HandlerInterceptor interceptor = this.interceptorList.get(i);try {interceptor.afterCompletion(request, response, this.handler, ex);} catch (Throwable ex2) {logger.error("HandlerInterceptor.afterCompletion threw exception", ex2);}}
}

倒序执行已执行过preHandle的拦截器。
异常场景3:
preHandle或者反射调用目标方法(Controller接口)过程中有异常抛出时，经过异常捕获将异常包装成Exception对象，并通过processDispatchResult方法处理结果：

protected void doDispatch(HttpServletRequest request, HttpServletResponse response) throws Exception {try {//。。。} catch (Exception ex) {dispatchException = ex;} catch (Throwable err) {dispatchException = new NestedServletException("Handler dispatch failed", err);}processDispatchResult(processedRequest, response, mappedHandler, mv, dispatchException);
}

processDispatchResult的主线逻辑如下：

private void processDispatchResult(HttpServletRequest request, HttpServletResponse response, @Nullable HandlerExecutionChain mappedHandler, @Nullable ModelAndView mv, @Nullable Exception exception) throws Exception {if (exception != null) {mv = processHandlerException(request, response, mappedHandler.getHandler(), exception);}mappedHandler.triggerAfterCompletion(request, response, null);
}

如果有异常抛出，则exception不为空，通过processHandlerException方法处理异常场景。
如论是否有异常抛出，最后都通过mappedHandler.triggerAfterCompletion会触发拦截的afterCompletion方法。
同样需要注意：拦截器在执行preHandle过程中抛出的异常，则会反向执行已执行过preHandle方法拦截器的afterCompletion方法。
如果在拦截器的postHandle或者反射调用目标方法过程中抛出的异常，则会执行所有拦截器的afterCompletion方法，流程如下所示：