之前已经就spring中的BeanPostProcessor使用方法以及其实现细节谈论过,现在从更加宏观、抽象的角度去理解spring的bpp具体是如何工作的,现在spring自身有多少bpp,如果我们有自定义的bpp需求,应该如何实现。
其中如下demo的代码位置 GitHub simple-spring
目录
再谈Spring BeanPostProcessor
1、BeanPostProcessor 种类
1.1、BeanPostProcessor 接口
1.2、InstantiationAwareBeanPostProcessor 接口
1.3、MergedBeanDefinitionPostProcessor 接口
2、源码学习
2.1、实例化前的提前处理
2.2、实例化后的合并
2.3、实例化后的数据填充
2.4、初始化init方法
3、Demo
1、BeanPostProcessor 种类
如上图,是在IDEA中使用control+H
命令显示出来的BeanPostProcessor的继承实现类关系图,重点介绍BeanPostProcessor
,InstantiationAwareBeanPostProcessor
、MergedBeanDefinitionPostProcessor
三个接口,后面再继续结合实际的实现类,分析spring是如何操作BeanPostProcessor对象的
阅读这里面的源码需要对Spring 源码有一定的理解,具体的可看【目录】Spring 源码学习
1.1、BeanPostProcessor 接口
public interface BeanPostProcessor {
Object postProcessBeforeInitialization(Object bean, String beanName) throws BeansException;
// 在调用bean的init方法(非构造方法)前调用
// beanName 是在Spring IOC 容器的bean 名称,返回的对象需要会被直接使用
// 切记!!!默认返回bean即可,不要无缘无故返回null,会出现各种NPE的情况
Object postProcessAfterInitialization(Object bean, String beanName) throws BeansException;
// 在调用bean的init方法(非构造方法)后调用
}
1.2、InstantiationAwareBeanPostProcessor 接口
public interface InstantiationAwareBeanPostProcessor extends BeanPostProcessor {
Object postProcessBeforeInstantiation(Class> beanClass, String beanName) throws BeansException;
// 在对象实例化之前被调用,传入的参数是类型以及bean的那么
// 如果有返回非空的对象,则意味着不需要调用doCreate操作完成对象实例化等
// 同时返回非空后会调用postProcessAfterInitialization方法
// 注意看,不是postProcessAfterInstantiation而是postProcessAfterInitialization方法
boolean postProcessAfterInstantiation(Object bean, String beanName) throws BeansException;
// 返回的是boolean类型,并不是和postProcessBeforeInstantiation配套使用的,在正常的实例化之后
// 主要功能是判断是否需要完成对象填充
// 如果返回false,则意味着不会完成对象属性的填充,例如@Resource导入的对象还是为null
PropertyValues postProcessPropertyValues(
PropertyValues pvs, PropertyDescriptor[] pds, Object bean, String beanName) throws BeansException;
// 提供给已经实例化的对象一种可以自定义完成对象pv属性的修改操作
// 注意不要随意返回null,一旦返回null,不会进行对象填充、对象依赖等操作
}
1.3、MergedBeanDefinitionPostProcessor 接口
public interface MergedBeanDefinitionPostProcessor extends BeanPostProcessor {
void postProcessMergedBeanDefinition(RootBeanDefinition beanDefinition, Class> beanType, String beanName);
// 在完成bean的实例化之后,填充数据(populateBean)之前,可自定义的修改beanDefinition内容
// 名字成为merge,合并BPP对象和对于的beanDefinition的内容
// 实际上可以完成任何想任何实现的功能
}
以上的三个接口基本上能够覆盖所有的BPP特定的功能点,从上面图也可以看出来,spring、springboot很多模块都有相应的对象实现,完成特定的功能。
2、源码学习
2.1、实例化前的提前处理
直接定位到AbstractAutowireCapableBeanFactory类的createBean方法,主要介绍的是InstantiationAwareBeanPostProcessor
try {
Object bean = resolveBeforeInstantiation(beanName, mbdToUse);
if (bean != null) {
// 如果bean有用,则会直接返回,不会再继续执行doCreateBean方法
return bean;
}
}
catch (Throwable ex) {
throw new BeanCreationException(mbdToUse.getResourceDescription(), beanName,
"BeanPostProcessor before instantiation of bean failed", ex);
// bean创建失败,在初始化之前错误了
}
protected Object resolveBeforeInstantiation(String beanName, RootBeanDefinition mbd) {
Object bean = null;
if (!Boolean.FALSE.equals(mbd.beforeInstantiationResolved)) {
// Make sure bean class is actually resolved at this point.
if (!mbd.isSynthetic() && hasInstantiationAwareBeanPostProcessors()) {
// 如果存在InstantiationAwareBeanPostProcessor对象
Class> targetType = determineTargetType(beanName, mbd);
if (targetType != null) {
bean = applyBeanPostProcessorsBeforeInstantiation(targetType, beanName);
// 调用的是postProcessBeforeInstantiation方法
if (bean != null) {
// 如果返回的数据不为null,则调用的是postProcessAfterInitialization方法
// 注意别看错了,一个是实例化Instantiation,一个是初始化Initialization
bean = applyBeanPostProcessorsAfterInitialization(bean, beanName);
}
}
}
mbd.beforeInstantiationResolved = (bean != null);
// 记录下是否通过提前批次的处理(如果bean不为null,则肯定是)
}
// 最后返回结果
return bean;
}
其意思就是利用在实例化之前检查是否存在合适的InstantiationAwareBeanPostProcessor对象,去拦截某些需要被处理的bean提前完成bean的实例化过程,不会去调用init方法,也没有数据的填充,@Resource对象的引入等操作。
2.2、实例化后的合并
实例化后,仅仅是完成了对象最基础的实例化工作,还未涉及到填充数据,init方法执行等操作,主要介绍的是MergedBeanDefinitionPostProcessor
synchronized (mbd.postProcessingLock) {
if (!mbd.postProcessed) {
try {
// 合并管理bean
applyMergedBeanDefinitionPostProcessors(mbd, beanType, beanName);
}
catch (Throwable ex) {
throw new BeanCreationException(mbd.getResourceDescription(), beanName,
"Post-processing of merged bean definition failed", ex);
}
mbd.postProcessed = true;
}
}
protected void applyMergedBeanDefinitionPostProcessors(RootBeanDefinition mbd, Class> beanType, String beanName) {
for (BeanPostProcessor bp : getBeanPostProcessors()) {
if (bp instanceof MergedBeanDefinitionPostProcessor) {
// 遍历所有的BeanPostProcessor
// 筛选出属于MergedBeanDefinitionPostProcessor的对象,调用postProcessMergedBeanDefinition
MergedBeanDefinitionPostProcessor bdp = (MergedBeanDefinitionPostProcessor) bp;
bdp.postProcessMergedBeanDefinition(mbd, beanType, beanName);
}
}
}
2.3、实例化后的数据填充
主要的操作在populateBean方法中,涉及到的也是InstantiationAwareBeanPostProcessor
再次强调InstantiationAwareBeanPostProcessor接口不仅仅是用在实例化前,在实例化之后也同样有用
boolean continueWithPropertyPopulation = true;
// 设置可以继续设置pv值的boolean对象为true
if (!mbd.isSynthetic() && hasInstantiationAwareBeanPostProcessors()) {
for (BeanPostProcessor bp : getBeanPostProcessors()) {
if (bp instanceof InstantiationAwareBeanPostProcessor) {
// 如果是InstantiationAwareBeanPostProcessor对象
InstantiationAwareBeanPostProcessor ibp = (InstantiationAwareBeanPostProcessor) bp;
if (!ibp.postProcessAfterInstantiation(bw.getWrappedInstance(), beanName)) {
// 调用postProcessAfterInstantiation
// 如果返回false意味着不需要填充pv操作了,直接break
continueWithPropertyPopulation = false;
break;
}
}
}
}
if (!continueWithPropertyPopulation) {
// 上文设置为false后,直接返回,完全忽略后续的applyPropertyValues操作
return;
}
if (hasInstAwareBpps || needsDepCheck) {
PropertyDescriptor[] filteredPds = filterPropertyDescriptorsForDependencyCheck(bw, mbd.allowCaching);
if (hasInstAwareBpps) {
for (BeanPostProcessor bp : getBeanPostProcessors()) {
if (bp instanceof InstantiationAwareBeanPostProcessor) {
InstantiationAwareBeanPostProcessor ibp = (InstantiationAwareBeanPostProcessor) bp;
pvs = ibp.postProcessPropertyValues(pvs, filteredPds, bw.getWrappedInstance(), beanName);
// 对bean对象的pv值进行处理操作
if (pvs == null) {
// 如果返回为null,则直接返回
// 所以在自定义设置BPP的时候必须注意该方法的返回值
// 通过ide自动生成的对象,其默认返回值是null
return;
}
}
}
}
if (needsDepCheck) {
// 检查对象的依赖问题
checkDependencies(beanName, mbd, filteredPds, pvs);
}
}
2.4、初始化init方法
在spring中是可以设置对象的init方法,在对象实现之后,在initializeBean方法中,如下的代码片段
Object wrappedBean = bean;
if (mbd == null || !mbd.isSynthetic()) {
wrappedBean = applyBeanPostProcessorsBeforeInitialization(wrappedBean, beanName);
// 调用postProcessBeforeInitialization
}
try {
invokeInitMethods(beanName, wrappedBean, mbd);
// 动态代理反射invoke调用初始化方法
// 就是在这里获取到bean的init方法调用信息,invoke调用
}
catch (Throwable ex) {
throw new BeanCreationException(
(mbd != null ? mbd.getResourceDescription() : null),
beanName, "Invocation of init method failed", ex);
}
if (mbd == null || !mbd.isSynthetic()) {
wrappedBean = applyBeanPostProcessorsAfterInitialization(wrappedBean, beanName);
// 唯一一个被调用了2次的地方的方法postProcessAfterInitialization
}
public Object applyBeanPostProcessorsBeforeInitialization(Object existingBean, String beanName)
throws BeansException {
Object result = existingBean;
for (BeanPostProcessor beanProcessor : getBeanPostProcessors()) {
result = beanProcessor.postProcessBeforeInitialization(result, beanName);
if (result == null) {
return result;
}
}
return result;
}
public Object applyBeanPostProcessorsAfterInitialization(Object existingBean, String beanName)
throws BeansException {
Object result = existingBean;
for (BeanPostProcessor beanProcessor : getBeanPostProcessors()) {
result = beanProcessor.postProcessAfterInitialization(result, beanName);
if (result == null) {
// 都有个这样的操作,返回为null,则退出
return result;
}
}
return result;
}
3、Demo
public class CustomInstantiationAwareBeanPostProcessor implements InstantiationAwareBeanPostProcessor {
public Object postProcessBeforeInstantiation(Class> beanClass, String beanName) throws BeansException {
if (beanName.equals("student")) {
Student student = new Student("bpp", 20);
return student;
}
if (beanName.equals("superStudent")) {
SuperStudent studentAndTeacher = new SuperStudent();
studentAndTeacher.setName("zhangsan");
return studentAndTeacher;
}
return null;
}
public boolean postProcessAfterInstantiation(Object bean, String beanName) throws BeansException {
return true;
}
public PropertyValues postProcessPropertyValues(PropertyValues pvs, PropertyDescriptor[] pds, Object bean, String beanName) throws BeansException {
//return pvs;
return null;
}
public Object postProcessBeforeInitialization(Object bean, String beanName) throws BeansException {
return bean;
}
public Object postProcessAfterInitialization(Object bean, String beanName) throws BeansException {
return bean;
}
}
public class SuperStudent {
private String name;
public String getName() {
return name;
}
public void setName(String name) {
this.name = name;
}
@Resource
private Teacher teacher;
public void doInit() {
System.out.println("superstudent init");
}
public void doSet() {
if (teacher != null) {
teacher.setAge(19);
teacher.setName("superStudent");
System.out.println(teacher.toString());
} else {
System.out.println("teacher is null");
}
}
@Override
public String toString() {
return "SuperStudent{" +
"name='" + name + '\'' +
", teacher=" + teacher +
'}';
}
}
public static void runBPP() {
ClassPathXmlApplicationContext applicationContext = new ClassPathXmlApplicationContext(
new String[]{"context-bpp.xml"});
SuperStudent student = (SuperStudent) applicationContext.getBean("superStudent");
Teacher teacher = (Teacher) applicationContext.getBean("teacher");
System.out.println(teacher.toString());
System.out.println(student.toString());
student.doSet();
System.out.println(teacher.toString());
}
如下图几种样例的代码
- 返回pv为null
很清楚在SuperStudent的对象是null,并没有完成填充操作,但是调用了init方法
- 返回pv为原值,不作任何改变
完成了填充操作,而且这个SuperStudent的teacher对象是和Spring IOC 容器内的保持一致
- 设置postProcessBeforeInstantiation
SuperStudent返回输出的对象,teacher没有值,而且也没有调用其init的方法