請求首先通過Filter chain，Filter主要包括ActionContextCleanUp，它主要清理當前線程的ActionContext和Dispatcher；FilterDispatcher主要通過AcionMapper來決定需要調用哪個Action。
ActionMapper取得了ActionMapping后，在Dispatcher的serviceAction方法里創建ActionProxy，ActionProxy創建ActionInvocation，然后ActionInvocation調用Interceptors，執行Action本身，創建Result并返回，當然，如果要在返回之前做些什么，可以實現PreResultListener。
ActionMapper其實是HttpServletRequest和Action調用請求的一個映射，它屏蔽了Action對于Request等java Servlet類的依賴。Struts2中它的默認實現類是DefaultActionMapper，ActionMapper很大的用處可以根據自己的需要來設計url格式，它自己也有Restful的實現，具體可以參考文檔的docs\actionmapper.html。
Action的一個代理，由ActionProxyFactory創建，它本身不包括Action實例，默認實現DefaultActionProxy是由ActionInvocation持有Action實例。ActionProxy作用是如何取得Action，無論是本地還是遠程。而ActionInvocation的作用是如何執行Action，攔截器的功能就是在ActionInvocation中實現的。
ConfigurationProvider就是Struts2中配置文件的解析器，Struts2中的配置文件主要是尤其實現類XmlConfigurationProvider及其子類StrutsXmlConfigurationProvider來解析，
1、客戶端發送請求
2、請求先通過ActionContextCleanUp-->FilterDispatcher
3、FilterDispatcher通過ActionMapper來決定這個Request需要調用哪個Action
4、如果ActionMapper決定調用某個Action，FilterDispatcher把請求的處理交給ActionProxy，這兒已經轉到它的Delegate--Dispatcher來執行
5、ActionProxy根據ActionMapping和ConfigurationManager找到需要調用的Action類
6、ActionProxy創建一個ActionInvocation的實例
7、ActionInvocation調用真正的Action，當然這涉及到相關攔截器的調用
8、Action執行完畢，ActionInvocation創建Result并返回，當然，如果要在返回之前做些什么，可以實現PreResultListener。添加PreResultListener可以在Interceptor中實現，不知道其它還有什么方式？
Struts2(2.1.2)部分源碼閱讀
從org.apache.struts2.dispatcher.FilterDispatcher開始
//創建Dispatcher，此類是一個Delegate，它是真正完成根據url解析，讀取對應Action的地方
public void init(FilterConfig filterConfig) throws ServletException {
try {
this.filterConfig = filterConfig;
initLogging();
dispatcher = createDispatcher(filterConfig);
dispatcher.init();
dispatcher.getContainer().inject(this);
//讀取初始參數pakages，調用parse()，解析成類似/org/apache/struts2/static,/template的數組
String param = filterConfig.getInitParameter("packages");
String packages = "org.apache.struts2.static template org.apache.struts2.interceptor.debugging";
if (param != null) {
packages = param + " " + packages;
}
this.pathPrefixes = parse(packages);
} finally {
ActionContext.setContext(null);
}
}
順著流程我們看Disptcher的init方法。init方法里就是初始讀取一些配置文件等，先看init_DefaultProperties，主要是讀取properties配置文件。
private void init_DefaultProperties() {
configurationManager.addConfigurationProvider(new DefaultPropertiesProvider());
}
打開DefaultPropertiesProvider
public void register(ContainerBuilder builder, LocatableProperties props)
throws ConfigurationException {
Settings defaultSettings = null;
try {
defaultSettings = new PropertiesSettings("org/apache/struts2/default");
} catch (Exception e) {
throw new ConfigurationException("Could not find or error in org/apache/struts2/default.properties", e);
}
loadSettings(props, defaultSettings);
}
//PropertiesSettings
//讀取org/apache/struts2/default.properties的配置信息，如果項目中需要覆蓋，可以在classpath里的struts.properties里覆寫
public PropertiesSettings(String name) {
URL settingsUrl = ClassLoaderUtils.getResource(name + ".properties", getClass());
if (settingsUrl == null) {
LOG.debug(name + ".properties missing");
settings = new LocatableProperties();
return;
}
settings = new LocatableProperties(new LocationImpl(null, settingsUrl.toString()));
// Load settings
InputStream in = null;
try {
in = settingsUrl.openStream();
settings.load(in);
} catch (IOException e) {
throw new StrutsException("Could not load " + name + ".properties:" + e, e);
} finally {
if(in != null) {
try {
in.close();
} catch(IOException io) {
LOG.warn("Unable to close input stream", io);
}
}
}
}
再來看init_TraditionalXmlConfigurations方法，這個是讀取struts-default.xml和Struts.xml的方法。
private void init_TraditionalXmlConfigurations() {
//首先讀取web.xml中的config初始參數值
//如果沒有配置就使用默認的"struts-default.xml,struts-plugin.xml,struts.xml"，
//這兒就可以看出為什么默認的配置文件必須取名為這三個名稱了
//如果不想使用默認的名稱，直接在web.xml中配置config初始參數即可
String configPaths = initParams.get("config");
if (configPaths == null) {
configPaths = DEFAULT_CONFIGURATION_PATHS;
}
String[] files = configPaths.split("\\s*[,]\\s*");
//依次解析配置文件，xwork.xml單獨解析
for (String file : files) {
if (file.endsWith(".xml")) {
if ("xwork.xml".equals(file)) {
configurationManager.addConfigurationProvider(new XmlConfigurationProvider(file, false));
} else {
configurationManager.addConfigurationProvider(new StrutsXmlConfigurationProvider(file, false, servletContext));
}
} else {
throw new IllegalArgumentException("Invalid configuration file name");
}
}
}
對于其它配置文件只用StrutsXmlConfigurationProvider，此類繼承XmlConfigurationProvider，而XmlConfigurationProvider又實現ConfigurationProvider接口。類XmlConfigurationProvider負責配置文件的讀取和解析，addAction()方法負責讀取<action>標簽，并將數據保存在ActionConfig中；addResultTypes()方法負責將<result-type>標簽轉化為ResultTypeConfig對象；loadInterceptors()方法負責將<interceptor>標簽轉化為InterceptorConfi對象；loadInterceptorStack()方法負責將<interceptor-ref>標簽轉化為InterceptorStackConfig對象；loadInterceptorStacks()方法負責將<interceptor-stack>標簽轉化成InterceptorStackConfig對象。而上面的方法最終會被addPackage()方法調用，將所讀取到的數據匯集到PackageConfig對象中。來看XmlConfigurationProvider的源代碼，詳細的我自己也就大體瀏覽了一下，各位可以自己研讀。
protected PackageConfig addPackage(Element packageElement) throws ConfigurationException {
PackageConfig.Builder newPackage = buildPackageContext(packageElement);
if (newPackage.isNeedsRefresh()) {
return newPackage.build();
}
.
addResultTypes(newPackage, packageElement);
loadInterceptors(newPackage, packageElement);
loadDefaultInterceptorRef(newPackage, packageElement);
loadDefaultClassRef(newPackage, packageElement);
loadGlobalResults(newPackage, packageElement);
loadGobalExceptionMappings(newPackage, packageElement);
NodeList actionList = packageElement.getElementsByTagName("action");
for (int i = 0; i < actionList.getLength(); i++) {
Element actionElement = (Element) actionList.item(i);
addAction(actionElement, newPackage);
}
loadDefaultActionRef(newPackage, packageElement);
PackageConfig cfg = newPackage.build();
configuration.addPackageConfig(cfg.getName(), cfg);
return cfg;
}
這兒發現一個配置上的小技巧，我的xwork2.0.*是沒有的，但是看源碼是看到xwork2.1.*是可以的。繼續看XmlConfigurationProvider的源代碼：
private List loadConfigurationFiles(String fileName, Element includeElement) {
List<Document> docs = new ArrayList<Document>();
if (!includedFileNames.contains(fileName)) {
Element rootElement = doc.getDocumentElement();
NodeList children = rootElement.getChildNodes();
int childSize = children.getLength();
for (int i = 0; i < childSize; i++) {
Node childNode = children.item(i);
if (childNode instanceof Element) {
Element child = (Element) childNode;
final String nodeName = child.getNodeName();
//解析每個action配置是，對于include文件可以使用通配符*來進行配置
//如Struts.xml中可配置成<include file="actions_*.xml"/>
if (nodeName.equals("include")) {
String includeFileName = child.getAttribute("file");
if(includeFileName.indexOf('*') != -1 ) {
ClassPathFinder wildcardFinder = new ClassPathFinder();
wildcardFinder.setPattern(includeFileName);
Vector<String> wildcardMatches = wildcardFinder.findMatches();
for (String match : wildcardMatches) {
docs.addAll(loadConfigurationFiles(match, child));
}
}
else {
docs.addAll(loadConfigurationFiles(includeFileName, child));
}
}
}
}
docs.add(doc);
loadedFileUrls.add(url.toString());
}
}
return docs;
}
init_CustomConfigurationProviders方式初始自定義的Provider，配置類全名和實現ConfigurationProvider接口，用逗號隔開即可。
private void init_CustomConfigurationProviders() {
String configProvs = initParams.get("configProviders");
if (configProvs != null) {
String[] classes = configProvs.split("\\s*[,]\\s*");
for (String cname : classes) {
try {
Class cls = ClassLoaderUtils.loadClass(cname, this.getClass());
ConfigurationProvider prov = (ConfigurationProvider)cls.newInstance();
configurationManager.addConfigurationProvider(prov);
}
}
}
}
好了，現在再回到FilterDispatcher，每次發送一個Request，FilterDispatcher都會調用doFilter方法。
public void doFilter(ServletRequest req, ServletResponse res, FilterChain chain) throws IOException, ServletException {
HttpServletRequest request = (HttpServletRequest) req;
HttpServletResponse response = (HttpServletResponse) res;
ServletContext servletContext = getServletContext();
String timerKey = "FilterDispatcher_doFilter: ";
try {
ValueStack stack = dispatcher.getContainer().getInstance(ValueStackFactory.class).createValueStack();
ActionContext ctx = new ActionContext(stack.getContext());
ActionContext.setContext(ctx);
UtilTimerStack.push(timerKey);
//根據content type來使用不同的Request封裝，可以參見Dispatcher的wrapRequest
request = prepareDispatcherAndWrapRequest(request, response);
ActionMapping mapping;
try {
//根據url取得對應的Action的配置信息--ActionMapping，actionMapper是通過Container的inject注入的
mapping = actionMapper.getMapping(request, dispatcher.getConfigurationManager());
} catch (Exception ex) {
log.error("error getting ActionMapping", ex);
dispatcher.sendError(request, response, servletContext, HttpServletResponse.SC_INTERNAL_SERVER_ERROR, ex);
return;
}
//如果找不到對應的action配置，則直接返回。比如你輸入***.jsp等等
//這兒有個例外，就是如果path是以“/struts”開頭，則到初始參數packages配置的包路徑去查找對應的靜態資源并輸出到頁面流中，當然.class文件除外。如果再沒有則跳轉到404
if (mapping == null) {
// there is no action in this request, should we look for a static resource?
String resourcePath = RequestUtils.getServletPath(request);
if ("".equals(resourcePath) && null != request.getPathInfo()) {
resourcePath = request.getPathInfo();
}
if (serveStatic && resourcePath.startsWith("/struts")) {
String name = resourcePath.substring("/struts".length());
findStaticResource(name, request, response);
} else {
chain.doFilter(request, response);
}
return;
}
//正式開始Action的方法了
dispatcher.serviceAction(request, response, servletContext, mapping);
} finally {
try {
ActionContextCleanUp.cleanUp(req);
} finally {
UtilTimerStack.pop(timerKey);
}
}
}
Dispatcher類的serviceAction方法：
public void serviceAction(HttpServletRequest request, HttpServletResponse response, ServletContext context,ActionMapping mapping) throws ServletException {
Map<String, Object> extraContext = createContextMap(request, response, mapping, context);
// If there was a previous value stack, then create a new copy and pass it in to be used by the new Action
ValueStack stack = (ValueStack) request.getAttribute(ServletActionContext.STRUTS_VALUESTACK_KEY);
if (stack != null) {
extraContext.put(ActionContext.VALUE_STACK, valueStackFactory.createValueStack(stack));
}
String timerKey = "Handling request from Dispatcher";
try {
UtilTimerStack.push(timerKey);
String namespace = mapping.getNamespace();
String name = mapping.getName();
String method = mapping.getMethod();
Configuration config = configurationManager.getConfiguration();
ActionProxy proxy = config.getContainer().getInstance(ActionProxyFactory.class).createActionProxy(
namespace, name, method, extraContext, true, false);
request.setAttribute(ServletActionContext.STRUTS_VALUESTACK_KEY, proxy.getInvocation().getStack());
// if the ActionMapping says to go straight to a result, do it!
if (mapping.getResult() != null) {
Result result = mapping.getResult();
result.execute(proxy.getInvocation());
} else {
proxy.execute();
}
// If there was a previous value stack then set it back onto the request
if (stack != null) {
request.setAttribute(ServletActionContext.STRUTS_VALUESTACK_KEY, stack);
}
} catch (ConfigurationException e) {
LOG.error("Could not find action or result", e);
sendError(request, response, context, HttpServletResponse.SC_NOT_FOUND, e);
} catch (Exception e) {
sendError(request, response, context, HttpServletResponse.SC_INTERNAL_SERVER_ERROR, e);
} finally {
UtilTimerStack.pop(timerKey);
}
}
第一句createContextMap()方法，該方法主要把Application、Session、Request的key value值拷貝到Map中，并放在HashMap<String,Object>中，可以參見createContextMap方法：
public Map<String,Object> createContextMap(HttpServletRequest request, HttpServletResponse response,
ActionMapping mapping, ServletContext context) {
// request map wrapping the http request objects
Map requestMap = new RequestMap(request);
// parameters map wrapping the http parameters. ActionMapping parameters are now handled and applied separately
Map params = new HashMap(request.getParameterMap());
// session map wrapping the http session
Map session = new SessionMap(request);
// application map wrapping the ServletContext
Map application = new ApplicationMap(context);
Map<String,Object> extraContext = createContextMap(requestMap, params, session, application, request, response, context);
extraContext.put(ServletActionContext.ACTION_MAPPING, mapping);
return extraContext;
}
后面才是最主要的--ActionProxy，ActionInvocation。ActionProxy是Action的一個代理類，也就是說Action的調用是通過ActionProxy實現的，其實就是調用了ActionProxy.execute()方法，而該方法又調用了ActionInvocation.invoke()方法。歸根到底，最后調用的是DefaultActionInvocation.invokeAction()方法。先看DefaultActionInvocation的init方法。
public void init(ActionProxy proxy) {
this.proxy = proxy;
Map contextMap = createContextMap();
// Setting this so that other classes, like object factories, can use the ActionProxy and other
// contextual information to operate
ActionContext actionContext = ActionContext.getContext();
if(actionContext != null) {
actionContext.setActionInvocation(this);
}
//創建Action，可Struts2里是每次請求都新建一個Action
createAction(contextMap);
if (pushAction) {
stack.push(action);
contextMap.put("action", action);
}
invocationContext = new ActionContext(contextMap);
invocationContext.setName(proxy.getActionName());
// get a new List so we don't get problems with the iterator if someone changes the list
List interceptorList = new ArrayList(proxy.getConfig().getInterceptors());
interceptors = interceptorList.iterator();
}
protected void createAction(Map contextMap) {
// load action
String timerKey = "actionCreate: "+proxy.getActionName();
try {
UtilTimerStack.push(timerKey);
//這兒默認建立Action是StrutsObjectFactory，實際中我使用的時候都是使用Spring創建的Action，這個時候使用的是SpringObjectFactory
action = objectFactory.buildAction(proxy.getActionName(), proxy.getNamespace(), proxy.getConfig(), contextMap);
}
..
} finally {
UtilTimerStack.pop(timerKey);
}
if (actionEventListener != null) {
action = actionEventListener.prepare(action, stack);
}
}
接下來看看DefaultActionInvocation 的invoke方法。
public String invoke() throws Exception {
String profileKey = "invoke: ";
try {
UtilTimerStack.push(profileKey);
if (executed) {
throw new IllegalStateException("Action has already executed");
}
//先執行interceptors
if (interceptors.hasNext()) {
final InterceptorMapping interceptor = (InterceptorMapping) interceptors.next();
UtilTimerStack.profile("interceptor: "+interceptor.getName(),
new UtilTimerStack.ProfilingBlock<String>() {
public String doProfiling() throws Exception {
resultCode = interceptor.getInterceptor().intercept(DefaultActionInvocation.this);
return null;
}
});
} else {
//interceptor執行完了之后執行action
resultCode = invokeActionOnly();
}
// this is needed because the result will be executed, then control will return to the Interceptor, which will
// return above and flow through again
if (!executed) {
//在Result返回之前調用preResultListeners
if (preResultListeners != null) {
for (Iterator iterator = preResultListeners.iterator();
iterator.hasNext();) {
PreResultListener listener = (PreResultListener) iterator.next();
String _profileKey="preResultListener: ";
try {
UtilTimerStack.push(_profileKey);
listener.beforeResult(this, resultCode);
}
finally {
UtilTimerStack.pop(_profileKey);
}
}
}
// now execute the result, if we're supposed to
if (proxy.getExecuteResult()) {
executeResult();
}
executed = true;
}
return resultCode;
}
finally {
UtilTimerStack.pop(profileKey);
}
}
看程序中的if(interceptors.hasNext())語句，當然，interceptors里存儲的是interceptorMapping列表（它包括一個Interceptor和一個name），所有的截攔器必須實現Interceptor的intercept方法，而該方法的參數恰恰又是ActionInvocation，在intercept方法中還是調用invocation.invoke()，從而實現了一個Interceptor鏈的調用。當所有的Interceptor執行完，最后調用invokeActionOnly方法來執行Action相應的方法。
protected String invokeAction(Object action, ActionConfig actionConfig) throws Exception {
String methodName = proxy.getMethod();
String timerKey = "invokeAction: "+proxy.getActionName();
try {
UtilTimerStack.push(timerKey);
boolean methodCalled = false;
Object methodResult = null;
Method method = null;
try {
//獲得需要執行的方法
method = getAction().getClass().getMethod(methodName, new Class[0]);
} catch (NoSuchMethodException e) {
//如果沒有對應的方法，則使用do+Xxxx來再次獲得方法
try {
String altMethodName = "do" + methodName.substring(0, 1).toUpperCase() + methodName.substring(1);
method = getAction().getClass().getMethod(altMethodName, new Class[0]);
} catch (NoSuchMethodException e1) {
// well, give the unknown handler a shot
if (unknownHandler != null) {
try {
methodResult = unknownHandler.handleUnknownActionMethod(action, methodName);
methodCalled = true;
} catch (NoSuchMethodException e2) {
// throw the original one
throw e;
}
} else {
throw e;
}
}
}
if (!methodCalled) {
methodResult = method.invoke(action, new Object[0]);
}
//根據不同的Result類型返回不同值
//如輸出流Result
if (methodResult instanceof Result) {
this.explicitResult = (Result) methodResult;
return null;
} else {
return (String) methodResult;
}
} catch (NoSuchMethodException e) {
throw new IllegalArgumentException("The " + methodName + "() is not defined in action " + getAction().getClass() + "");
} catch (InvocationTargetException e) {
// We try to return the source exception.
Throwable t = e.getTargetException();
if (actionEventListener != null) {
String result = actionEventListener.handleException(t, getStack());
if (result != null) {
return result;
}
}
if (t instanceof Exception) {
throw(Exception) t;
} else {
throw e;
}
} finally {
UtilTimerStack.pop(timerKey);
}
}
好了，action執行完了，還要根據ResultConfig返回到view，也就是在invoke方法中調用executeResult方法。
private void executeResult() throws Exception {
//根據ResultConfig創建Result
result = createResult();
String timerKey = "executeResult: "+getResultCode();
try {
UtilTimerStack.push(timerKey);
if (result != null) {
//這兒正式執行:)
//可以參考Result的實現，如用了比較多的ServletDispatcherResult,ServletActionRedirectResult,ServletRedirectResult
result.execute(this);
} else if (resultCode != null && !Action.NONE.equals(resultCode)) {
throw new ConfigurationException("No result defined for action " + getAction().getClass().getName()
+ " and result " + getResultCode(), proxy.getConfig());
} else {
if (LOG.isDebugEnabled()) {
LOG.debug("No result returned for action "+getAction().getClass().getName()+" at "+proxy.getConfig().getLocation());
}
}
} finally {
UtilTimerStack.pop(timerKey);
}
}
public Result createResult() throws Exception {
if (explicitResult != null) {
Result ret = explicitResult;
explicitResult = null;;
return ret;
}
ActionConfig config = proxy.getConfig();
Map results = config.getResults();
ResultConfig resultConfig = null;
synchronized (config) {
try {
//根據result名稱獲得ResultConfig，resultCode就是result的name
resultConfig = (ResultConfig) results.get(resultCode);
} catch (NullPointerException e) {
}
if (resultConfig == null) {
//如果找不到對應name的ResultConfig，則使用name為*的Result
resultConfig = (ResultConfig) results.get("*");
}
}
if (resultConfig != null) {
try {
//參照StrutsObjectFactory的代碼
Result result = objectFactory.buildResult(resultConfig, invocationContext.getContextMap());
return result;
} catch (Exception e) {
LOG.error("There was an exception while instantiating the result of type " + resultConfig.getClassName(), e);
throw new XWorkException(e, resultConfig);
}
} else if (resultCode != null && !Action.NONE.equals(resultCode) && unknownHandler != null) {
return unknownHandler.handleUnknownResult(invocationContext, proxy.getActionName(), proxy.getConfig(), resultCode);
}
return null;
}
//StrutsObjectFactory
public Result buildResult(ResultConfig resultConfig, Map extraContext) throws Exception {
String resultClassName = resultConfig.getClassName();
if (resultClassName == null)
return null;
//創建Result，因為Result是有狀態的，所以每次請求都新建一個
Object result = buildBean(resultClassName, extraContext);
//這句很重要，后面將會談到，reflectionProvider參見OgnlReflectionProvider；
//resultConfig.getParams()就是result配置文件里所配置的參數<param></param>
//setProperties方法最終調用的是Ognl類的setValue方法
//這句其實就是把param名值設置到根對象result上
reflectionProvider.setProperties(resultConfig.getParams(), result, extraContext);
if (result instanceof Result)
return (Result) result;
throw new ConfigurationException(result.getClass().getName() + " does not implement Result.");
}
public class TestOgnl {
private User user;
private Map context;
@Before
public void setUp() throws Exception {
}
@Test
public void ognlGetValue() throws Exception {
reset();
Assert.assertEquals("myyate", Ognl.getValue("name", user));
Assert.assertEquals("cares", Ognl.getValue("dept.name", user));
Assert.assertEquals("myyate", Ognl.getValue("name", context, user));
Assert.assertEquals("contextmap", Ognl.getValue("#name", context, user));
Assert.assertEquals("parker", Ognl.getValue("#pen", context, user));
}
@Test
public void ognlSetValue() throws Exception {
reset();
Ognl.setValue("name", user, "myyateC");
Assert.assertEquals("myyateC", Ognl.getValue("name", user));
Ognl.setValue("dept.name", user, "caresC");
Assert.assertEquals("caresC", Ognl.getValue("dept.name", user));
Assert.assertEquals("contextmap", Ognl.getValue("#name", context, user));
Ognl.setValue("#name", context, user, "contextmapC");
Assert.assertEquals("contextmapC", Ognl.getValue("#name", context, user));
Assert.assertEquals("parker", Ognl.getValue("#pen", context, user));
Ognl.setValue("#name", context, user, "parkerC");
Assert.assertEquals("parkerC", Ognl.getValue("#name", context, user));
}
public static void main(String[] args) throws Exception {
JUnitCore.runClasses(TestOgnl.class);
}
private void reset() {
user = new User("myyate", new Dept("cares"));
context = new OgnlContext();
context.put("pen", "parker");
context.put("name", "contextmap");
}
}
class User {
public User(String name, Dept dept) {
this.name = name;
this.dept = dept;
}
String name;
private Dept dept;
public Dept getDept() {
return dept;
}
public String getName() {
return name;
}
public void setDept(Dept dept) {
this.dept = dept;
}
public void setName(String name) {
this.name = name;
}
}
class Dept {
public Dept(String name) {
this.name = name;
}
private String name;
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
}
這樣，一個Struts2的請求流程基本上就結束了。其實我覺得做項目把Struts2參考文檔看兩遍就可以了，呵呵！（寫博客比看代碼還累）