前提
(1)作者技术比较差,文章写的比较随意,也可能有错误,欢迎您指出。
(2)如果您不了解Spring Bean的声明周期,那么您可以看一下文章(Bean的生命周期)或者百度其它文章,然后在回来看该文章,否则个人感觉应该看不懂
解决循环依赖
假设有一种下面的情况,A中有B,B中有A
@Data
public class A {
private B b;
public A() {System.out.println("A 无参构造器。。。");}
public void speak() {System.out.println("------AAA---------");}
}
@Data
public class B {
public B() {System.out.println("B 无参构造器。。。");}
private A a;
public void speak() {System.out.println("------BBB---------");}
}
图片分析
代码分析
在创建的A的时候调用doCreateBean方法
1)调用A无参构造方法创建Bean
2)把该bean对象添加到三级缓存中(在下面的代码中有注释)
3)Bean的属性赋值,A的里面引用了B,所以此时会调用doCteateBean(B)
//AbstractAutowireCapableBeanFactory
protected Object doCreateBean(final String beanName, final RootBeanDefinition mbd, final @Nullable Object[] args)
throws BeanCreationException {
{
// Instantiate the bean.
BeanWrapper instanceWrapper = null;
if (mbd.isSingleton()) {
//省略
}
if (instanceWrapper == null) {
//1)调用无参构造方法创建Bean
instanceWrapper = createBeanInstance(beanName, mbd, args);
}
final Object bean = instanceWrapper.getWrappedInstance();
Class<?> beanType = instanceWrapper.getWrappedClass();
if (beanType != NullBean.class) {
mbd.resolvedTargetType = beanType;
}
synchronized (mbd.postProcessingLock) {
//省略
}
boolean earlySingletonExposure = (mbd.isSingleton() && this.allowCircularReferences &&
isSingletonCurrentlyInCreation(beanName));
if (earlySingletonExposure) {
if (logger.isTraceEnabled()) {
//省略
}
//2)把该bean对象添加到三级缓存中,注意getEarlyBeanReference方法,特别有用
addSingletonFactory(beanName, () -> getEarlyBeanReference(beanName, mbd, bean));
}
// Initialize the bean instance.
Object exposedObject = bean;
try {
//3)Bean的属性赋值
populateBean(beanName, mbd, instanceWrapper);
//4)处理aware接口、applyBeanPostProcessorsBeforeInitialization、initMethod
exposedObject = initializeBean(beanName, exposedObject, mbd);
}
catch (Throwable ex) {
if (ex instanceof BeanCreationException && beanName.equals(((BeanCreationException) ex).getBeanName())) {
//省略
}
else {
//省略
}
}
if (earlySingletonExposure) {
//省略
}
// Register bean as disposable.
try {
registerDisposableBeanIfNecessary(beanName, bean, mbd);
}
catch (BeanDefinitionValidationException ex) {
//省略
}
return exposedObject;
}
}
此时在创建B的时候调用getBean(A),然后会走到下面代码的地方,从三级缓存中获取到A(B=null),返回该不完整的A的地址,然后B创建成功,然后继续创建A,然后A也创建成功。
-------------------------源码1
//DefaultSingletonBeanRegistry
protected Object getSingleton(String beanName, boolean allowEarlyReference) {
//从一级缓存中获取,即IOC容器,即完整的Bean对象
Object singletonObject = this.singletonObjects.get(beanName);
if (singletonObject == null && isSingletonCurrentlyInCreation(beanName)) {
synchronized (this.singletonObjects) {
//从二级缓存中获取
singletonObject = this.earlySingletonObjects.get(beanName);
if (singletonObject == null && allowEarlyReference) {
//从三级缓存中获取
ObjectFactory<?> singletonFactory = this.singletonFactories.get(beanName);
if (singletonFactory != null) {
singletonObject = singletonFactory.getObject();
this.earlySingletonObjects.put(beanName, singletonObject);
this.singletonFactories.remove(beanName);
}
}
}
}
return singletonObject;
}
循环依赖总结
(1)创建A的时候调用A的无参构造方法,然后在把得到的地址A(B=null)放入到三级缓存中,然后填充自己的属性B,也就会创建B;
(2)当创建B的时候,填充自己的属性A,从三级缓存中拿到A(B=null)地址,然后B创建成功;
(3)此时回到(1),此时拿到B,然后完善A,创建A成功。
(4)因为在(2)中拿到的是A的地址,所以在(3)中完善A在B中是一个。
三级缓存
疑问
个人感觉二级缓存足矣,为什么还要三级缓存?
反驳疑问
假设下面的场景:只有singletonObject(第一级缓存)和singletonFactory (第三级缓存),即没有earlySingletonObjects(第二级缓存)
如果有这么一种情况A(B),B(A),还有一个AOP是关注A的某个方法
此时的逻辑为:
1)创建A
2)把A(B=null)的地址(abc)存入singletonFactory缓存中
3)创建B
4)B在赋值a属性的时候,在singletonFactory缓存中拿出A的地址(abc)并且赋值给属性a(左边这句话是错的)(这就是三级缓存的关键),
4.1)没有AOP的时候,确实是存的a的地址,没错,返回的也是a的地址。
4.2)如果有AOP,确实存进去的是a的地址,但是返回的已经不是A的地址了,是A的代理对象地址(看源码2,3,4)。
总结:此时就出现问题了,如果没有earlySingletonObjects(第二级缓存),那么每次在singletonFactory (第三级缓存)中拿到的A对象都会创建创建一个代理对象,即每次向依赖A的对象中赋的值都是不同的代理对象,那么就不符合单例模式了。
-------------------------源码2
protected Object getEarlyBeanReference(String beanName, RootBeanDefinition mbd, Object bean) {
Object exposedObject = bean;
if (!mbd.isSynthetic() && hasInstantiationAwareBeanPostProcessors()) {
for (BeanPostProcessor bp : getBeanPostProcessors()) {
if (bp instanceof SmartInstantiationAwareBeanPostProcessor) {
SmartInstantiationAwareBeanPostProcessor ibp = (SmartInstantiationAwareBeanPostProcessor) bp;
exposedObject = ibp.getEarlyBeanReference(exposedObject, beanName);
}
}
}
return exposedObject;
}
-------------------------源码3
//AbstractAutoProxyCreator
@Override
public Object getEarlyBeanReference(Object bean, String beanName) {
Object cacheKey = getCacheKey(bean.getClass(), beanName);
this.earlyProxyReferences.put(cacheKey, bean);
//跟进去
return wrapIfNecessary(bean, beanName, cacheKey);
}
-------------------------源码4
//AbstractAutoProxyCreator
protected Object wrapIfNecessary(Object bean, String beanName, Object cacheKey) {
if (StringUtils.hasLength(beanName) && this.targetSourcedBeans.contains(beanName)) {
return bean;
}
if (Boolean.FALSE.equals(this.advisedBeans.get(cacheKey))) {
return bean;
}
if (isInfrastructureClass(bean.getClass()) || shouldSkip(bean.getClass(), beanName)) {
this.advisedBeans.put(cacheKey, Boolean.FALSE);
return bean;
}
// Create proxy if we have advice.
Object[] specificInterceptors = getAdvicesAndAdvisorsForBean(bean.getClass(), beanName, null);
if (specificInterceptors != DO_NOT_PROXY) {
this.advisedBeans.put(cacheKey, Boolean.TRUE);
//返回了一个新对象,新地址
//返回了一个新对象,新地址
//返回了一个新对象,新地址
Object proxy = createProxy(
bean.getClass(), beanName, specificInterceptors, new SingletonTargetSource(bean));
this.proxyTypes.put(cacheKey, proxy.getClass());
return proxy;
}
this.advisedBeans.put(cacheKey, Boolean.FALSE);
return bean;
}
总结
1)在没有AOP的情况下二级缓存足矣解决循环依赖,三级缓存更能解决问题。
2)三级缓存其实也是解决循环依赖的,是解决带AOP的循环依赖的,如上文中举的例子。如果您查的三级缓存资料没有说AOP,个人感觉这篇文章写的不是很充实。
本文没有回答的疑问
疑问1
上问中反驳二级缓存不能解决带AOP的循环依赖问题时,是把earlySingletonObjects(第二级缓存)去掉;如果我说我去掉singletonFactory (第三级缓存),那该如何反驳二级缓存不能解决带AOP的循环依赖问题呢???
疑问2
就拿上问中举的例字来说,A依赖B,B依赖A,有一个关注A的AOP。
下面是创建Bean声明周期的一段代码,以创建A为例
//AbstractAutowireCapableBeanFactory
protected Object doCreateBean{
//创建A
Object exposedObject = bean;
try {
//初始化A,因为A中有属性B,此时去创建B,然后把A的代理对象存入earlySingletonObjects缓存中,B创建完毕,然后又回到此处继续初始化A
populateBean(beanName, mbd, instanceWrapper);
//为非代理对象A执行aware接口等等
exposedObject = initializeBean(beanName, exposedObject, mbd);
}
catch (Throwable ex) {
//省略
}
}
if (earlySingletonExposure) {
//在earlySingletonObjects中拿到代理对象A
Object earlySingletonReference = getSingleton(beanName, false);
if (earlySingletonReference != null) {
if (exposedObject == bean) {
//把exposedObject由指向非代理对象A变为指向代理对象A,那么
//exposedObject = initializeBean(beanName, exposedObject, mbd);
//我认为是白做了,我不清楚这个地方???????????????
exposedObject = earlySingletonReference;
}
else if (!this.allowRawInjectionDespiteWrapping && hasDependentBean(beanName)) {
String[] dependentBeans = getDependentBeans(beanName);
Set<String> actualDependentBeans = new LinkedHashSet<>(dependentBeans.length);
for (String dependentBean : dependentBeans) {
if (!removeSingletonIfCreatedForTypeCheckOnly(dependentBean)) {
actualDependentBeans.add(dependentBean);
}
}
if (!actualDependentBeans.isEmpty()) {
throw new BeanCurrentlyInCreationException(beanName,
"Bean with name '" + beanName + "' has been injected into other beans [" +
StringUtils.collectionToCommaDelimitedString(actualDependentBeans) +
"] in its raw version as part of a circular reference, but has eventually been " +
"wrapped. This means that said other beans do not use the final version of the " +
"bean. This is often the result of over-eager type matching - consider using " +
"'getBeanNamesOfType' with the 'allowEagerInit' flag turned off, for example.");
}
}
}
}