sitemesh3是一个页面布局框架,可以是你的页面无缝的嵌入到一个整体页面中,从而是你的页面看起来风格是一致的,所有的一切从这里开始
ConfigurableSiteMeshFilter
public void init(FilterConfig filterConfig) throws ServletException {
this.filterConfig = filterConfig;
configProperties = getConfigProperties(filterConfig);
autoReload = getAutoReload();
logger.config("Auto reloading " + (autoReload ? "enabled" : "disabled"));
synchronized (configLock) {
deployNewFilter(setup());
}
initialized = true;
}
public void doFilter(ServletRequest servletRequest, ServletResponse servletResponse, FilterChain filterChain) throws IOException, ServletException {
if (!initialized) {
throw new ServletException(getClass().getName() + ".init(FilterConfig) was not called");
}
reloadIfNecessary();
filter.doFilter(servletRequest, servletResponse, filterChain);
}
sitemesh3是从该filter开始的,首先看init(),了解它初始化过程
我们看到configProperties = getConfigProperties(filterConfig);,是filter的配置信息,autoReload = getAutoReload(); 查看配置文件是否需要自动重加载,默认是开启的。从后边可以知道它是通过对比sitemesh3.xml的最后修改时间来确定是不是需要加载的
这些只是准备工作,关键点在 deployNewFilter(setup()); ,也就是setup方法中
setup会创建一个filter,当执行到doFilter() ,内部实际执行的就是filter.doFilter(),到这里就可以肯定所有核心内容都在这个filter中,包括init和filter的过程,首先我们看init的过程,了解它都需要那些东西,然后再看doFilter中的内容,来决定我们怎么改进它。
protected Filter setup() throws ServletException {
ObjectFactory objectFactory = getObjectFactory();
SiteMeshFilterBuilder builder = new SiteMeshFilterBuilder();
new PropertiesFilterConfigurator(objectFactory, configProperties)
.configureFilter(builder);
new XmlFilterConfigurator(getObjectFactory(),
loadConfigXml(filterConfig, getConfigFileName()))
.configureFilter(builder);
applyCustomConfiguration(builder);
return builder.create();
}
一行一行的来讲ObjectFactory里面其实只有一行代码 return class.forName(className) ,就是为类生成对象,同时隔离到那些exception。
builder 这个类很重要,所有的配置信息都会存到里面,PropertiesFilterConfigurator 这行是将web.xml 中sitemeshfilter中的配置信息,存放到builder中
下面一行是将sitemesh3.xml中的配置信息,存放到builder中。
applyCustomConfiguration(builder);这行是为了通过java编码的方式来配置filter
配置信息有 exclude, decorate, minetype,content-processor,tag-rule-bundle
exclude 是不需要装饰,decorate是需要装饰的, minetype 是默认返回的页面类型, content-processor 是自定义的页面内容处理器,tag-rule-bundle 是自定义的标签,如<sitemesh:write/>等。
这里就可以看清楚sitemesh配置有三种方式 1,filter,2,sitemesh.xml, 3 java硬编码 ,混合使用的话,优先级从搞到低
下面我们去详细看看builder中有什么东西
Filter create() {
return new SiteMeshFilter(
getSelector(),
getContentProcessor(),
getDecoratorSelector());
}
看到没有一个热乎的filter出炉了, 用到了builder的三个属性, selector 选择器,contentProcessor 内容处理器,decoratorSelector 装饰选择器
我们继续深入:selector 是关于mineType决定是否拦截 , contentProcessor 是解析标签的 ,decoratorSelector 是关于装饰页面和被装饰页面的关系
好了,初始化过程基本就看完了,接下来就到正餐时间了,从上面我们没有看到更多的默认配置信息, 下一步就是这里filter.doFilter() ,等把这个过程理完了,然后再把代码通读一遍就能明白了
SiteMeshFilter.class
FilterConfig filterConfig = getFilterConfig();
if (filterConfig == null) {
// TODO: Is this really necessary? Can we survive without init() being called?
throw new ServletException(getClass().getName() + ".init() has not been called.");
}
HttpServletRequest request = (HttpServletRequest) servletRequest;
HttpServletResponse response = (HttpServletResponse) servletResponse;
ServletContext servletContext = filterConfig.getServletContext();
if (!selector.shouldBufferForRequest(request)) {
// Optimization: If the content doesn't need to be buffered,
// skip the rest of this filter.
filterChain.doFilter(request, response);
return;
}
if (containerTweaks.shouldAutoCreateSession()) {
// Some containers (such as Tomcat 4) will not allow sessions
// to be created in the decorator. (i.e after the
// response has been committed).
request.getSession(true);
}
try {
// The main work.
bufferAndPostProcess(filterChain, request, response, selector);
} catch (IllegalStateException e) {
// Some containers (such as WebLogic) throw an IllegalStateException when an error page is served.
// It may be ok to ignore this. However, for safety it is propegated if possible.
if (!containerTweaks.shouldIgnoreIllegalStateExceptionOnErrorPage()) {
throw e;
}
} catch (RuntimeException e) {
if (containerTweaks.shouldLogUnhandledExceptions()) {
// Some containers (such as Tomcat 4) swallow RuntimeExceptions in filters.
servletContext.log("Unhandled exception occurred whilst decorating page", e);
}
throw e;
}
FilterConfig filterConfig = getFilterConfig(); 还记得这是什么时候注入的吗
!selector.shouldBufferForRequest(request) 初步判断: 是否已经呗sitemesh filter拦截过了, 是否不需要装饰
if (containerTweaks.shouldAutoCreateSession()) { 是否返回false, 是为适应tomcat4 写的
接下来看 关键部分 bufferAndPostProcess ,前面几行可以不用看 关键在于:
。。。。
filterChain.doFilter(wrapRequest(request), responseBuffer);
CharBuffer buffer = responseBuffer.getBuffer();
// If content was buffered, post-process it.
boolean processed = false;
if (buffer != null && !responseBuffer.bufferingWasDisabled()) {
processed = postProcess(responseBuffer.getContentType(), buffer, request, response, metaData);
}
if (!response.isCommitted()) {
responseBuffer.preCommit();
}
// If no decoratoes applied, and we have some buffered content, write the original.
if (buffer != null && !processed) {
writeOriginal(response, buffer, responseBuffer);
}
省略的几行是封装response,以便得到返回的html内容,还有就是控制response 的状态 如 if_modified_since 等
buffer是 返回的页面内容,已经被web容器解析过了。
在postProcess 开始进行装饰,如果没有成功就就buffer直接写入到response中
看postProcess 方法:
WebAppContext context = createContext(contentType, request, response, metaData);
Content content = contentProcessor.build(buffer, context);
if (content == null) {
return false;
}
String[] decoratorPaths = decoratorSelector.selectDecoratorPaths(content, context);
for (String decoratorPath : decoratorPaths) {
content = context.decorate(decoratorPath, content);
}
if (content == null) {
return false;
}
content.getData().writeValueTo(response.getWriter());
return true;
首先用rule规则将内容解析一遍,再找到要装饰的页面,然后用dispatch去解析,取得内容后,用context将被装饰页面和装饰页面一起解析
接下来我们着重看看一下 contentProccesor.build() 和 context.decorate() 和 context.build() 即可
就先到这里吧,自己一边看一边写挺累的