bean生命周期
实例化 Instantiation >>>> createBeanInstance(beanName,mbd,args)
属性赋值 Populate >>>> populateBean(beanName,mbd,isntanceWrapper)
初始化 Initialization >>>> initialzeBean(beanName,exposeObject,mbd)
其中前三个逻辑都在doCreateBean方法中
protected Object doCreateBean(final String beanName, final RootBeanDefinition mbd, final @Nullable Object[] args)
throws BeanCreationException {
// Instantiate the bean.
BeanWrapper instanceWrapper = null;
if (mbd.isSingleton()) {
instanceWrapper = this.factoryBeanInstanceCache.remove(beanName);
}
if (instanceWrapper == null) {
//实例化(构造方法反射创建)
instanceWrapper = createBeanInstance(beanName, mbd, args);
}
//无关代码
// Eagerly cache singletons to be able to resolve circular references
// even when triggered by lifecycle interfaces like BeanFactoryAware.
boolean earlySingletonExposure = (mbd.isSingleton() && this.allowCircularReferences &&
isSingletonCurrentlyInCreation(beanName));
if (earlySingletonExposure) {
if (logger.isDebugEnabled()) {
logger.debug("Eagerly caching bean '" + beanName +
"' to allow for resolving potential circular references");
}
//放入第三级缓存中,提前暴露,解决循环引用
addSingletonFactory(beanName, () -> getEarlyBeanReference(beanName, mbd, bean));
}
// Initialize the bean instance.
Object exposedObject = bean;
try {
//属性填充
populateBean(beanName, mbd, instanceWrapper);
//初始化
exposedObject = initializeBean(beanName, exposedObject, mbd);
}
//无关代码
return exposedObject;
}
首先在实例化之前,spring会扫描所有添加@Sercvice\@Controller\@Component\@Repository的类
将他们封装成BeanDefinition对象(用来描述类的定义信息),其源码如下:
/**
* A BeanDefinition describes a bean instance, which has property values,
* constructor argument values, and further information supplied by
* concrete implementations.
*
* <p>This is just a minimal interface: The main intention is to allow a
* {@link BeanFactoryPostProcessor} such as {@link PropertyPlaceholderConfigurer}
* to introspect and modify property values and other bean metadata.
*
* @author Juergen Hoeller
* @author Rob Harrop
* @since 19.03.2004
* @see ConfigurableListableBeanFactory#getBeanDefinition
* @see org.springframework.beans.factory.support.RootBeanDefinition
* @see org.springframework.beans.factory.support.ChildBeanDefinition
*/
public interface BeanDefinition extends AttributeAccessor, BeanMetadataElement {
/**
* Scope identifier for the standard singleton scope: "singleton".
* <p>Note that extended bean factories might support further scopes.
* @see #setScope
*/
String SCOPE_SINGLETON = ConfigurableBeanFactory.SCOPE_SINGLETON;
/**
* Scope identifier for the standard prototype scope: "prototype".
* <p>Note that extended bean factories might support further scopes.
* @see #setScope
*/
String SCOPE_PROTOTYPE = ConfigurableBeanFactory.SCOPE_PROTOTYPE;
/**
* Role hint indicating that a {@code BeanDefinition} is a major part
* of the application. Typically corresponds to a user-defined bean.
*/
int ROLE_APPLICATION = 0;
/**
* Role hint indicating that a {@code BeanDefinition} is a supporting
* part of some larger configuration, typically an outer
* {@link org.springframework.beans.factory.parsing.ComponentDefinition}.
* {@code SUPPORT} beans are considered important enough to be aware
* of when looking more closely at a particular
* {@link org.springframework.beans.factory.parsing.ComponentDefinition},
* but not when looking at the overall configuration of an application.
*/
int ROLE_SUPPORT = 1;
/**
* Role hint indicating that a {@code BeanDefinition} is providing an
* entirely background role and has no relevance to the end-user. This hint is
* used when registering beans that are completely part of the internal workings
* of a {@link org.springframework.beans.factory.parsing.ComponentDefinition}.
*/
int ROLE_INFRASTRUCTURE = 2;
// Modifiable attributes
/**
* Set the name of the parent definition of this bean definition, if any.
*/
void setParentName(@Nullable String parentName);
/**
* Return the name of the parent definition of this bean definition, if any.
*/
@Nullable
String getParentName();
/**
* Specify the bean class name of this bean definition.
* <p>The class name can be modified during bean factory post-processing,
* typically replacing the original class name with a parsed variant of it.
* @see #setParentName
* @see #setFactoryBeanName
* @see #setFactoryMethodName
*/
void setBeanClassName(@Nullable String beanClassName);
/**
* Return the current bean class name of this bean definition.
* <p>Note that this does not have to be the actual class name used at runtime, in
* case of a child definition overriding/inheriting the class name from its parent.
* Also, this may just be the class that a factory method is called on, or it may
* even be empty in case of a factory bean reference that a method is called on.
* Hence, do <i>not</i> consider this to be the definitive bean type at runtime but
* rather only use it for parsing purposes at the individual bean definition level.
* @see #getParentName()
* @see #getFactoryBeanName()
* @see #getFactoryMethodName()
*/
@Nullable
String getBeanClassName();
/**
* Override the target scope of this bean, specifying a new scope name.
* @see #SCOPE_SINGLETON
* @see #SCOPE_PROTOTYPE
*/
void setScope(@Nullable String scope);
/**
* Return the name of the current target scope for this bean,
* or {@code null} if not known yet.
*/
@Nullable
String getScope();
/**
* Set whether this bean should be lazily initialized.
* <p>If {@code false}, the bean will get instantiated on startup by bean
* factories that perform eager initialization of singletons.
*/
void setLazyInit(boolean lazyInit);
/**
* Return whether this bean should be lazily initialized, i.e. not
* eagerly instantiated on startup. Only applicable to a singleton bean.
*/
boolean isLazyInit();
/**
* Set the names of the beans that this bean depends on being initialized.
* The bean factory will guarantee that these beans get initialized first.
*/
void setDependsOn(@Nullable String... dependsOn);
/**
* Return the bean names that this bean depends on.
*/
@Nullable
String[] getDependsOn();
/**
* Set whether this bean is a candidate for getting autowired into some other bean.
* <p>Note that this flag is designed to only affect type-based autowiring.
* It does not affect explicit references by name, which will get resolved even
* if the specified bean is not marked as an autowire candidate. As a consequence,
* autowiring by name will nevertheless inject a bean if the name matches.
*/
void setAutowireCandidate(boolean autowireCandidate);
/**
* Return whether this bean is a candidate for getting autowired into some other bean.
*/
boolean isAutowireCandidate();
/**
* Set whether this bean is a primary autowire candidate.
* <p>If this value is {@code true} for exactly one bean among multiple
* matching candidates, it will serve as a tie-breaker.
*/
void setPrimary(boolean primary);
/**
* Return whether this bean is a primary autowire candidate.
*/
boolean isPrimary();
/**
* Specify the factory bean to use, if any.
* This the name of the bean to call the specified factory method on.
* @see #setFactoryMethodName
*/
void setFactoryBeanName(@Nullable String factoryBeanName);
/**
* Return the factory bean name, if any.
*/
@Nullable
String getFactoryBeanName();
/**
* Specify a factory method, if any. This method will be invoked with
* constructor arguments, or with no arguments if none are specified.
* The method will be invoked on the specified factory bean, if any,
* or otherwise as a static method on the local bean class.
* @see #setFactoryBeanName
* @see #setBeanClassName
*/
void setFactoryMethodName(@Nullable String factoryMethodName);
/**
* Return a factory method, if any.
*/
@Nullable
String getFactoryMethodName();
/**
* Return the constructor argument values for this bean.
* <p>The returned instance can be modified during bean factory post-processing.
* @return the ConstructorArgumentValues object (never {@code null})
*/
ConstructorArgumentValues getConstructorArgumentValues();
/**
* Return if there are constructor argument values defined for this bean.
* @since 5.0.2
*/
default boolean hasConstructorArgumentValues() {
return !getConstructorArgumentValues().isEmpty();
}
/**
* Return the property values to be applied to a new instance of the bean.
* <p>The returned instance can be modified during bean factory post-processing.
* @return the MutablePropertyValues object (never {@code null})
*/
MutablePropertyValues getPropertyValues();
/**
* Return if there are property values values defined for this bean.
* @since 5.0.2
*/
default boolean hasPropertyValues() {
return !getPropertyValues().isEmpty();
}
// Read-only attributes
/**
* Return whether this a <b>Singleton</b>, with a single, shared instance
* returned on all calls.
* @see #SCOPE_SINGLETON
*/
boolean isSingleton();
/**
* Return whether this a <b>Prototype</b>, with an independent instance
* returned for each call.
* @since 3.0
* @see #SCOPE_PROTOTYPE
*/
boolean isPrototype();
/**
* Return whether this bean is "abstract", that is, not meant to be instantiated.
*/
boolean isAbstract();
/**
* Get the role hint for this {@code BeanDefinition}. The role hint
* provides the frameworks as well as tools with an indication of
* the role and importance of a particular {@code BeanDefinition}.
* @see #ROLE_APPLICATION
* @see #ROLE_SUPPORT
* @see #ROLE_INFRASTRUCTURE
*/
int getRole();
/**
* Return a human-readable description of this bean definition.
*/
@Nullable
String getDescription();
/**
* Return a description of the resource that this bean definition
* came from (for the purpose of showing context in case of errors).
*/
@Nullable
String getResourceDescription();
/**
* Return the originating BeanDefinition, or {@code null} if none.
* Allows for retrieving the decorated bean definition, if any.
* <p>Note that this method returns the immediate originator. Iterate through the
* originator chain to find the original BeanDefinition as defined by the user.
*/
@Nullable
BeanDefinition getOriginatingBeanDefinition();
}
而获取到这些类的信息之后,spring会把他们存入到一个map中,方便获取BeanDefinition对象来获取类的信息
/** Map of bean definition objects, keyed by bean name */
private final Map<String, BeanDefinition> beanDefinitionMap = new ConcurrentHashMap<>(256);
/** List of bean definition names, in registration order */
private volatile List<String> beanDefinitionNames = new ArrayList<>(256);
以上为存储beanDefinition的map和beanDefinitionName的list
方法调用链:
通过BeanUtils创建对象,至此,对象就已经被创建出来了,然后到了这一步
protected Object doCreateBean(final String beanName, final RootBeanDefinition mbd, final @Nullable Object[] args)
throws BeanCreationException {
//实例化
// Eagerly cache singletons to be able to resolve circular references
// even when triggered by lifecycle interfaces like BeanFactoryAware.
boolean earlySingletonExposure = (mbd.isSingleton() && this.allowCircularReferences &&
isSingletonCurrentlyInCreation(beanName));
if (earlySingletonExposure) {
if (logger.isDebugEnabled()) {
logger.debug("Eagerly caching bean '" + beanName +
"' to allow for resolving potential circular references");
}
//加入第三级缓存,提前暴露来解决循环引用
addSingletonFactory(beanName, () -> getEarlyBeanReference(beanName, mbd, bean));
}
//属性填充
//初始化
return exposedObject;
}
跟踪进去
三层缓存的定义
/** Cache of singleton objects: bean name --> bean instance */
//spring容器/单例池/一级缓存
private final Map<String, Object> singletonObjects = new ConcurrentHashMap<>(256);
/** Cache of singleton factories: bean name --> ObjectFactory */
//三级缓存,用于循环引用
private final Map<String, ObjectFactory<?>> singletonFactories = new HashMap<>(16);
/** Cache of early singleton objects: bean name --> bean instance */
//二级缓存,用于用户拓展
private final Map<String, Object> earlySingletonObjects = new HashMap<>(16);
发现其将刚实例化好的对象放入到第三级缓存中,作用在后面会讲到。假如现在这个对象是personService,而他引用了userService,userService又引用了personService,这样就形成了循环引用,如下:
@Component
public class PersonService {
@Autowired
UserService userService;
public PersonService() {
System.out.println("PersonService");
}
}
@Component
public class UserService {
@Autowired
PersonService personService;
public UserService() {
System.out.println("UserService");
}
}
那么在接下来的属性填充中他是怎么做的呢
因为我们是用@Autowired进行自动注入,所以获取到AutowiredAnnotationBeanPostProcessor调用其postProcessPropertyValues进行属性后置处理
1.查找自动注入属性元数据
2.注入数据
到beanFactory中去拿userService
一路调用,getSingleton为null
因为它先去一级缓存singletonObjets(单例池)中去拿,拿不到,为null,并且是否创建为false,所以直接返回null,所以接下来开始了userService的生命周期,当其一路走到属性注入时(已经把自己放入到第三级缓存中)
又走到这步,通过beanFactory来解析此desc
可以看到最终走到getSingleton方法中,一级缓存没有获取到personService,又因为在创建userService的时候,在getSingleton重载方法中
执行了beforeSingleCreation
可以看到他将beanName放入到一个set集合中,标识当前单例对象正在创建中,而在上面getSingleton,有一处判断,是否正在创建当中,正是此作用,因此进入到二级缓存中去拿,拿不到,并且判断是否允许循环引用,而传值的时候就是true,所以到第三级缓存中去拿,最终拿到了(前面讲到personService讲自己放入到了三级缓存,提前暴露,而此时供userService获取,当然能获取到了),并且移除第三级缓存中的转而放入到第二级缓存
拿到需要注入的实例之后,一路返回,最后通过反射把值注入进去,到此userService已经实例化完毕
可以看到personService中的userService已经有值了,而personService中的userService还是null
然后一路返回到getSingleton方法中,执行afterSingletonCreation,里面逻辑就是从正在创建的set集合中移除userService,然后执行addSingleton方法,将经历完整生命周期的bean放入到单例池(spring容器)中
可以看到二三级缓存移除,放入一级缓存singletonObjects中,然后一路返回到personService属性赋值中
可以看到此时已经userService已经有值了
然后返回到inject方法中,熟悉的场景,通过反射给属性赋值 ,接下来和前面所述一样,最终会把经历完整生命周期的bean放入到单例池(spring容器)中。为什么构造器循环引用无法解决呢,因为在实例化对象的前提是执行构造方法。