1. 簡(jiǎn)單認(rèn)識(shí)BeanProcessorBeanProcessor的理解

BeanProcessor是spring中的一個(gè)重要接口，他有兩個(gè)接口方法一個(gè)是postProcessBeforeInitialization前置初始化，另一個(gè)是postProcessAfterInitialization后置初始化。從名稱上就可以大概清楚這個(gè)接口的作用：在一個(gè)業(yè)務(wù)流程的前后加入兩個(gè)接口方法，當(dāng)執(zhí)行這個(gè)業(yè)務(wù)流程時(shí)，就會(huì)觸發(fā)這兩個(gè)接口方法的執(zhí)行。簡(jiǎn)單的總結(jié)一下有兩個(gè)要點(diǎn)：

在業(yè)務(wù)流程中，根據(jù)BeanProcessor接口方法加在不同的位置(一般是前后)，可以實(shí)現(xiàn)對(duì)業(yè)務(wù)邏輯的擴(kuò)展。 在業(yè)務(wù)邏輯執(zhí)行前，BeanProcessor的實(shí)現(xiàn)類必須已經(jīng)被創(chuàng)建完成(BeanProcessor接口類必須要優(yōu)先實(shí)例化)。

而在spring中，就有很多實(shí)現(xiàn)了BeanProcessor的bean，通過(guò)在重要的業(yè)務(wù)流程(如bean的生命周期流程)的前后加上BeanProcessor接口方法，就可以對(duì)業(yè)務(wù)邏輯進(jìn)行修改或補(bǔ)充。

一個(gè)BeanProcessor的使用實(shí)例

在spring的bean生命周期中，BeanProcessor接口方法會(huì)在bean創(chuàng)建后的初始化方法(init-method或@PostConstruct指向的方法)前后執(zhí)行before和after方法；那有沒有在bean創(chuàng)建前后執(zhí)行的接口方法呢？答案是肯定有的，這個(gè)功能是由BeanProcessor的子接口InstantiationAwareBeanPostProcessor來(lái)實(shí)現(xiàn)的，他也是有before和after方法，會(huì)在bean實(shí)例化前后執(zhí)行。

我們先定義一個(gè)BeanProcessor接口實(shí)現(xiàn)類和一個(gè)InstantiationAwareBeanPostProcessor接口實(shí)現(xiàn)類。

BeanPostProcessor實(shí)現(xiàn)類:

//net.postProcessor.CustomerPostProcessor@Componentpublic class CustomerPostProcessor implements BeanPostProcessor { @PostConstruct public void init(){ System.out.println('執(zhí)行CustomerPostProcessor的PostConstruct'); } public CustomerPostProcessor(){ System.out.println('執(zhí)行CustomerPostProcessor的構(gòu)造方法'); } @Override public Object postProcessBeforeInitialization(Object bean, String beanName) throws BeansException { System.out.println(bean+'======BeforeInitialization======'+ beanName); return bean; } @Override public Object postProcessAfterInitialization(Object bean, String beanName) throws BeansException { System.out.println(bean+'======AfterInitialization======'+ beanName); return bean; }}

InstantiationAwareBeanPostProcessor實(shí)現(xiàn)類:

//net.postProcessor.CustomerInitialPostProcessor@Componentpublic class CustomerInitialPostProcessor implements InstantiationAwareBeanPostProcessor { @PostConstruct public void init(){ System.out.println('執(zhí)行CustomerInitialPostProcessor的PostConstruct'); } public CustomerInitialPostProcessor(){ System.out.println('執(zhí)行CustomerInitialPostProcessor的構(gòu)造方法'); } @Override public Object postProcessBeforeInstantiation(Class<?> beanClass, String beanName) throws BeansException { System.out.println('bean初始化前執(zhí)行：class為'+beanClass.getName()+'|beanName為'+beanName); return null; } @Override public boolean postProcessAfterInstantiation(Object bean, String beanName) throws BeansException { System.out.println('bean初始化后執(zhí)行：Object為'+bean+'|beanName為'+beanName); return false; }}

再創(chuàng)建一個(gè)普通的bean對(duì)象：

//net.postProcessor.FirstBean@Componentpublic class FirstBean implements InitializingBean { private String msg = 'hello'; @PostConstruct public void init(){ System.out.println('執(zhí)行FirstBean的PostConstruct'); } public FirstBean(){ System.out.println('FirstBean構(gòu)造方法！'+msg); } public String getMsg() { return msg; } public void setMsg(String msg) { this.msg = msg; } @Override public void afterPropertiesSet() throws Exception { System.out.println('執(zhí)行FirstBean的afterPropertiesSet'); }}

我們創(chuàng)建一個(gè)spring工廠對(duì)象將上述bean加載進(jìn)去：

@Testpublic void test(){ AnnotationConfigApplicationContext applicationContext = new AnnotationConfigApplicationContext('net.postProcessor');}//執(zhí)行得到以下結(jié)果：執(zhí)行CustomerInitialPostProcessor的構(gòu)造方法執(zhí)行CustomerInitialPostProcessor的PostConstruct執(zhí)行CustomerPostProcessor的構(gòu)造方法執(zhí)行CustomerPostProcessor的PostConstruct bean初始化前執(zhí)行：class為net.postProcessor.FirstBean|beanName為firstBeanFirstBean構(gòu)造方法！hellobean初始化后執(zhí)行：Object為net.postProcessor.FirstBean@79179359|beanName為firstBean net.postProcessor.FirstBean@79179359======BeforeInitialization======firstBean執(zhí)行FirstBean的PostConstruct執(zhí)行FirstBean的afterPropertiesSetnet.postProcessor.FirstBean@79179359======AfterInitialization======firstBean

通過(guò)上述結(jié)果證明了我們之前的說(shuō)法是正確的：

1.BeanPostProcessor接口類會(huì)優(yōu)先實(shí)例化，且在實(shí)例化中無(wú)法不會(huì)調(diào)用BeanPostProcessor接口方法的

2.InstantiationAwareBeanPostProcessor接口方法會(huì)在FirstBean構(gòu)造方法構(gòu)造方法前后執(zhí)行

3.BeanPostProcessor接口方法會(huì)在FirstBean實(shí)例化后進(jìn)行初始化的前后執(zhí)行

注意：若@PostConstruct注解方法方法未執(zhí)行，請(qǐng)加入javax.annotation:javax.annotation-api:1.3.2jar包依賴，原因是@PostConstruct是J2EE標(biāo)準(zhǔn)的注解，不是spring自己的接口，而在JDK8往上的版本中設(shè)計(jì)者打算棄用這些注解，所以做了處理，我們是沒有辦法直接使用J2EE標(biāo)準(zhǔn)注解的(@Resource、@PostConstruct、@PreDestroy等幾個(gè)注解)，為了兼容這種情況，所以有了javax.annotation-apijar包的產(chǎn)生(或者降低JDK版本)。

2. BeanProcessor的實(shí)現(xiàn)思路和簡(jiǎn)化實(shí)例BeanProcessor大概的實(shí)現(xiàn)思路

通過(guò)之前的了解BeanProcessor的使用，我們可以知道BeanProcessor并不復(fù)雜，但是卻十分的重要，下面來(lái)分析下BeanProcessor的實(shí)現(xiàn)思路：

創(chuàng)建個(gè)接口A，接口包含一些切點(diǎn)方法(Before、After、Around之類的)，實(shí)現(xiàn)這個(gè)接口A的類要在使用前就創(chuàng)建好 我們需要有個(gè)業(yè)務(wù)流程，這個(gè)業(yè)務(wù)流程由若干步組成；將接口A的接口方法插入到這些業(yè)務(wù)步驟之間(需要擴(kuò)展的地方) 要執(zhí)行這個(gè)業(yè)務(wù)流程時(shí)，把接口A的實(shí)現(xiàn)類對(duì)象賦值到業(yè)務(wù)流程中，在執(zhí)行業(yè)務(wù)流程中，就會(huì)觸發(fā)接口方法的執(zhí)行完成功能擴(kuò)展

當(dāng)我們更換賦值到業(yè)務(wù)流程中的接口A的實(shí)現(xiàn)類時(shí)，對(duì)應(yīng)的擴(kuò)展邏輯也會(huì)隨之變化，這樣就實(shí)現(xiàn)了可插拔式的擴(kuò)展邏輯(策略模式)。

一個(gè)BeanProcessor的簡(jiǎn)化邏輯實(shí)例

在spring中我們可以創(chuàng)建任意數(shù)量的bean實(shí)現(xiàn)BeanProcessor接口，所以實(shí)際上我們是要一個(gè)全局的beanProcessorList對(duì)象用來(lái)存儲(chǔ)這些BeanProcessor對(duì)象；在執(zhí)行業(yè)務(wù)代碼時(shí)，要循環(huán)這個(gè)beanProcessorList對(duì)象，獲取你需要的BeanProcessor對(duì)象來(lái)執(zhí)行接口方法。下面是一個(gè)模擬spring bean生命周期的簡(jiǎn)化版，來(lái)幫助你理解spring中BeanProcessor的工作原理。

net.postProcessor.SecondBean.java

@Componentpublic class SecondBean { private String msg = 'world'; public SecondBean(){ System.out.println('SecondBean構(gòu)造方法！'+msg); } public String getMsg() { return msg; } public void setMsg(String msg) { this.msg = msg; }}

net.postProcessor.CustomerPostProcessor.java

@Componentpublic class CustomerPostProcessor implements BeanPostProcessor { @PostConstruct public void init(){ System.out.println('執(zhí)行CustomerPostProcessor的PostConstruct'); } public CustomerPostProcessor(){ System.out.println('執(zhí)行CustomerPostProcessor的構(gòu)造方法'); } @Override public Object postProcessBeforeInitialization(Object bean, String beanName) throws BeansException { System.out.println(bean+'======BeforeInitialization======'+ beanName); return bean; } @Override public Object postProcessAfterInitialization(Object bean, String beanName) throws BeansException { System.out.println(bean+'======AfterInitialization======'+ beanName); return bean; }}net.postProcessor.PostProcessor.java

public class PostProcessor { //模擬掃描到的bean信息<'SecondBean', 'net.postProcessor.SecondBean'> Map<String, String> scanBeanMap = new HashMap<>(); //模擬spring的beanPostProcessors列表 List<BeanPostProcessor> processorBeanList = new ArrayList<>(); //模擬bean對(duì)象緩存 Map<String, Object> beanCache = new HashMap<>(); //添加掃描的bean信息 public PostProcessor addBeanInfo(String beanName, String classPath){ this.scanBeanMap.put(beanName, classPath); return this; } //模擬bean創(chuàng)建流程 public Object execute(){ try { //先臨時(shí)存儲(chǔ)實(shí)現(xiàn)了postProcessor接口的bean對(duì)象 List<BeanPostProcessor> postProcessorStrList = new ArrayList<>(); //循環(huán)scanBeanMap，獲取bean列表中實(shí)現(xiàn)了postProcessor接口的類，加入processorBeanList中 for(String temp: scanBeanMap.keySet()){ Class<?> clazz = Class.forName(scanBeanMap.get(temp)); //判斷是否實(shí)現(xiàn)了BeanPostProcessor接口 if(BeanPostProcessor.class.isAssignableFrom(clazz)){//實(shí)例化讓如臨時(shí)容器postProcessorStrList.add((BeanPostProcessor)createBean(temp)); } } //將實(shí)現(xiàn)了postProcessor接口的bean加入processorBeanList中 for(BeanPostProcessor obj: postProcessorStrList){ processorBeanList.add(obj); } //再次循環(huán)scanBeanMap初始化所用bean for(String temp: scanBeanMap.keySet()){ createBean(temp); } } catch (ClassNotFoundException e) { e.printStackTrace(); } return null; } //bean實(shí)例化 public Object createBean(String beanName){ //從緩存中獲取 if(beanCache.containsKey(beanName)){ return beanCache.get(beanName); }else{ //緩存中取不到，則進(jìn)行創(chuàng)建后加入緩存 try { Class<?> clazz = Class.forName(scanBeanMap.get(beanName)); //processor前置方法執(zhí)行 for(BeanPostProcessor processor : processorBeanList){processor.postProcessBeforeInitialization(clazz, beanName); } //bean實(shí)例化 Object result = clazz.getConstructor().newInstance(); //processor后置方法執(zhí)行 for(BeanPostProcessor processor : processorBeanList){processor.postProcessAfterInitialization(result, beanName); } //將bean加入緩存 beanCache.put(beanName, result); return result; } catch (ClassNotFoundException e) { e.printStackTrace(); } catch (IllegalAccessException e) { e.printStackTrace(); } catch (InstantiationException e) { e.printStackTrace(); } catch (NoSuchMethodException e) { e.printStackTrace(); } catch (InvocationTargetException e){ e.printStackTrace(); } } return null; }}

代碼調(diào)用

public static void main(String[] args) { PostProcessor postProcessor = new PostProcessor(); //添加掃描到的bean postProcessor .addBeanInfo('SecondBean', 'net.postProcessor.SecondBean') .addBeanInfo('CustomerPostProcessor', 'net.postProcessor.CustomerPostProcessor'); postProcessor.execute();}//執(zhí)行結(jié)果執(zhí)行CustomerPostProcessor的構(gòu)造方法class net.postProcessor.SecondBean======BeforeInitialization======SecondBeanSecondBean構(gòu)造方法！worldnet.postProcessor.SecondBean@1b40d5f0======AfterInitialization======SecondBean

代碼邏輯如下：

循環(huán)bean信息列表，將BeanPostProcessor接口bean分離出來(lái)優(yōu)先實(shí)例化(實(shí)例化中緩存bean對(duì)象)，并將之放入臨時(shí)容器。 循環(huán)完成，將臨時(shí)容器中的BeanPostProcessor接口bean賦值到全局BeanPostProcessor接口列表中 再次循環(huán)bean信息列表，緩存存在則直接返回緩存對(duì)象，不存在則進(jìn)行bean實(shí)例化，期間循環(huán)調(diào)用全局BeanPostProcessor接口對(duì)象方法3. spring中BeanProcessor的源碼解析

我們要從spring中的refresh()開始看起：

public void refresh() throws BeansException, IllegalStateException { synchronized (this.startupShutdownMonitor) { // Prepare this context for refreshing. //刷新準(zhǔn)備 prepareRefresh(); // Tell the subclass to refresh the internal bean factory. //告訴子類刷新內(nèi)部bean工廠。 ConfigurableListableBeanFactory beanFactory = obtainFreshBeanFactory(); // Prepare the bean factory for use in this context. //為容器準(zhǔn)備bean工程 prepareBeanFactory(beanFactory); try { // Allows post-processing of the bean factory in context subclasses. //允許在上下文bean的后處理工廠子類。 postProcessBeanFactory(beanFactory); // Invoke factory processors registered as beans in the context. //優(yōu)先將BeanDefinitionRegistryPostProcessorBeanFactoryPostProcessor接口的bean對(duì)象實(shí)例化 //屬于spring內(nèi)部組件調(diào)用 invokeBeanFactoryPostProcessors(beanFactory); // Register bean processors that intercept bean creation. //處理用戶自定義PostProcessor接口對(duì)象，之后加入spring的beanPostProcessors列表， // 供之后預(yù)實(shí)例化其他bean時(shí)觸發(fā)這些PostProcessor方法 registerBeanPostProcessors(beanFactory);//...省略代碼 //實(shí)例化所有(non-lazy-init)單件。finishBeanFactoryInitialization(beanFactory); } catch (BeansException ex) { if (logger.isWarnEnabled()) { logger.warn('Exception encountered during context initialization - ' + 'cancelling refresh attempt: ' + ex); } // Destroy already created singletons to avoid dangling resources. //bean銷毀 destroyBeans(); // Reset ’active’ flag. //取消刷新 cancelRefresh(ex); // Propagate exception to caller. throw ex; } finally { // Reset common introspection caches in Spring’s core, since we // might not ever need metadata for singleton beans anymore... //重置公共緩存 resetCommonCaches(); } }}

其中包含有postProcess字段都有可能和BeanProcessor相關(guān)，這里有三個(gè)相關(guān)方法：

postProcessBeanFactory(beanFactory)，這個(gè)是一共空的擴(kuò)展方法，顯然無(wú)關(guān) invokeBeanFactoryPostProcessors(beanFactory)，處理spring中實(shí)現(xiàn)了BeanProcessor接口的內(nèi)部組件直接調(diào)用接口方法 registerBeanPostProcessors(beanFactory)，實(shí)例化用戶自定義BeanProcessor接口bean組件，之后循環(huán)賦值到全局BeanProcessor列表中

所以registerBeanPostProcessors()就是我們要找的對(duì)象，來(lái)跟進(jìn)看下registerBeanPostProcessors()：

//AbstractApplicationContext#registerBeanPostProcessorsprotected void registerBeanPostProcessors(ConfigurableListableBeanFactory beanFactory) { //委托給PostProcessorRegistrationDelegate.registerBeanPostProcessors進(jìn)行處理PostProcessorRegistrationDelegate.registerBeanPostProcessors進(jìn)行處理(beanFactory, this);}

public static void registerBeanPostProcessors( ConfigurableListableBeanFactory beanFactory, AbstractApplicationContext applicationContext) { //查詢實(shí)現(xiàn)了BeanPostProcessor接口的beanName String[] postProcessorNames = beanFactory.getBeanNamesForType(BeanPostProcessor.class, true, false); // Register BeanPostProcessorChecker that logs an info message when // a bean is created during BeanPostProcessor instantiation, i.e. when // a bean is not eligible for getting processed by all BeanPostProcessors. int beanProcessorTargetCount = beanFactory.getBeanPostProcessorCount() + 1 + postProcessorNames.length; beanFactory.addBeanPostProcessor(new BeanPostProcessorChecker(beanFactory, beanProcessorTargetCount)); // Separate between BeanPostProcessors that implement PriorityOrdered, // Ordered, and the rest. List<BeanPostProcessor> priorityOrderedPostProcessors = new ArrayList<>(); List<BeanPostProcessor> internalPostProcessors = new ArrayList<>(); List<String> orderedPostProcessorNames = new ArrayList<>(); List<String> nonOrderedPostProcessorNames = new ArrayList<>(); //根據(jù)beanName循環(huán)調(diào)用getBean進(jìn)行實(shí)例化 for (String ppName : postProcessorNames) { if (beanFactory.isTypeMatch(ppName, PriorityOrdered.class)) { BeanPostProcessor pp = beanFactory.getBean(ppName, BeanPostProcessor.class); priorityOrderedPostProcessors.add(pp); if (pp instanceof MergedBeanDefinitionPostProcessor) { internalPostProcessors.add(pp); } } else if (beanFactory.isTypeMatch(ppName, Ordered.class)) { orderedPostProcessorNames.add(ppName); } else { nonOrderedPostProcessorNames.add(ppName); } } // First, register the BeanPostProcessors that implement PriorityOrdered. //對(duì)BeanPostProcessor接口對(duì)象進(jìn)行排序 sortPostProcessors(priorityOrderedPostProcessors, beanFactory); //將獲取到的PostProcessors接口對(duì)象加入到spring的beanPostProcessors列表 registerBeanPostProcessors(beanFactory, priorityOrderedPostProcessors); // Next, register the BeanPostProcessors that implement Ordered. List<BeanPostProcessor> orderedPostProcessors = new ArrayList<>(); for (String ppName : orderedPostProcessorNames) { BeanPostProcessor pp = beanFactory.getBean(ppName, BeanPostProcessor.class); orderedPostProcessors.add(pp); if (pp instanceof MergedBeanDefinitionPostProcessor) { internalPostProcessors.add(pp); } } sortPostProcessors(orderedPostProcessors, beanFactory); registerBeanPostProcessors(beanFactory, orderedPostProcessors); // Now, register all regular BeanPostProcessors. List<BeanPostProcessor> nonOrderedPostProcessors = new ArrayList<>(); for (String ppName : nonOrderedPostProcessorNames) { BeanPostProcessor pp = beanFactory.getBean(ppName, BeanPostProcessor.class); nonOrderedPostProcessors.add(pp); if (pp instanceof MergedBeanDefinitionPostProcessor) { internalPostProcessors.add(pp); } } registerBeanPostProcessors(beanFactory, nonOrderedPostProcessors); // Finally, re-register all internal BeanPostProcessors. sortPostProcessors(internalPostProcessors, beanFactory); registerBeanPostProcessors(beanFactory, internalPostProcessors); // Re-register post-processor for detecting inner beans as ApplicationListeners, // moving it to the end of the processor chain (for picking up proxies etc). beanFactory.addBeanPostProcessor(new ApplicationListenerDetector(applicationContext));}

果然這里就是處理BeanPostProcessor接口的地方，邏輯和之前的思路類似：

循環(huán)掃描到的bean列表，獲取實(shí)現(xiàn)了BeanPostProcessor接口的beanName數(shù)組 循環(huán)beanName數(shù)組數(shù)組，調(diào)用beanFactory.getBean()將bean實(shí)例化，并放入priorityOrderedPostProcessors列表中 調(diào)用sortPostProcessors對(duì)priorityOrderedPostProcessors列表進(jìn)行排序(處理BeanPostProcessor調(diào)用的順序) 調(diào)用registerBeanPostProcessors將priorityOrderedPostProcessors列表中的bean對(duì)象賦值到全局列表beanPostProcessors中 回到refresh()中，當(dāng)調(diào)用finishBeanFactoryInitialization()對(duì)所用bean進(jìn)行預(yù)實(shí)例化時(shí)就會(huì)調(diào)用這些BeanPostProcessor接口方法

以上就是spring BeanProcessor接口詳解的詳細(xì)內(nèi)容，更多關(guān)于spring BeanProcessor接口的資料請(qǐng)關(guān)注好吧啦網(wǎng)其它相關(guān)文章！