对于加载bean的功能,在Spring中的调用方式为:
MyTestBean bean = (MyTestBean)bf.getBean("myTestBean")
这行代码在Spring中是这样实现的
public T getBean(String name, Class requiredType, Object... args) throws BeansException {
return this.doGetBean(name, requiredType, args, false);
}
protected T doGetBean(String name, Class requiredType, final Object[] args, boolean typeCheckOnly) throws BeansException {
//提取对应的beanName
final String beanName = this.transformedBeanName(name);
/*
*检查缓存中或者实例工厂中是否有对应的实例
*为什么首先会使用这段代码呢,因为在创建单例bean的时候会存在依赖注入的情况,而在创建依赖的时候为了避免循环依赖,
*Spring创建bean的原则是不等bean创建完成就会将创建bean的ObjectFactory提早曝光,
*也就是将ObjectFactory加入到缓存中,一旦下个bean创建时候需要依赖上个bean则直接使用ObjectFactory
*/
//直接尝试从缓存获取或者singletonFactories中的ObjectFactory中获取
Object sharedInstance = this.getSingleton(beanName);
Object bean;
if(sharedInstance != null && args == null) {
if(this.logger.isDebugEnabled()) {
if(this.isSingletonCurrentlyInCreation(beanName)) {
this.logger.debug("Returning eagerly cached instance of singleton bean \'" + beanName + "\' that is not fully initialized yet - a consequence of a circular reference");
} else {
this.logger.debug("Returning cached instance of singleton bean \'" + beanName + "\'");
}
}
//返回对应的实例,有时候存在诸如BeanFactory的情况并不是直接返回实例本身而是返回指定方法返回的实例
bean = this.getObjectForBeanInstance(sharedInstance, name, beanName, (RootBeanDefinition)null);
} else {
//只有在单例情况下才会尝试解决循环依赖,原型模式情况下,如果存在
//A中有B的属性,B中有A的属性,那么当依赖注入的时候,就会产生当A还未创建完的时候因为
//对于B的创建再次返回创建A,造成循环依赖,也就是下面的情况
//isPrototypeCurrentlyInCreation(beanName)为true
if(this.isPrototypeCurrentlyInCreation(beanName)) {
throw new BeanCurrentlyInCreationException(beanName);
}
BeanFactory parentBeanFactory = this.getParentBeanFactory();
//如果BeanDefinitionMap中也就是在所有已经加载的类中不包括beanName则尝试从parentBeanFactory中检测
if(parentBeanFactory != null && !this.containsBeanDefinition(beanName)) {
String var19 = this.originalBeanName(name);
//递归到BeanFactory中寻找
if(args != null) {
return parentBeanFactory.getBean(var19, args);
}
return parentBeanFactory.getBean(var19, requiredType);
}
//如果不是仅仅做类型检查则是创建bean,这里要进行记录
if(!typeCheckOnly) {
this.markBeanAsCreated(beanName);
}
//将存储XML配置文件的GernericBeanDefinition转换为RootBeanDefinition,如果指定BeanName是子Bean的话同时会合并父类的相关属性
final RootBeanDefinition mbd = this.getMergedLocalBeanDefinition(beanName);
this.checkMergedBeanDefinition(mbd, beanName, args);
String[] dependsOn = mbd.getDependsOn();
String scopeName;
//若存在依赖则需要递归实例化依赖的bean
if(dependsOn != null) {
String[] var14 = dependsOn;
int ex = dependsOn.length;
for(int scope = 0; scope < ex; ++scope) {
scopeName = var14[scope];
this.getBean(scopeName);
//缓存依赖调用
this.registerDependentBean(scopeName, beanName);
}
}
//实例化依赖的bean后便可以实例化mbd本身了
//singleton模式的创建
if(mbd.isSingleton()) {
sharedInstance = this.getSingleton(beanName, new ObjectFactory() {
public Object getObject() throws BeansException {
try {
return AbstractBeanFactory.this.createBean(beanName, mbd, args);
} catch (BeansException var2) {
AbstractBeanFactory.this.destroySingleton(beanName);
throw var2;
}
}
});
bean = this.getObjectForBeanInstance(sharedInstance, name, beanName, mbd);
} else if(mbd.isPrototype()) {
//prototype模式的创建(new)
scopeName = null;
Object var20;
try {
this.beforePrototypeCreation(beanName);
var20 = this.createBean(beanName, mbd, args);
} finally {
this.afterPrototypeCreation(beanName);
}
bean = this.getObjectForBeanInstance(var20, name, beanName, mbd);
} else {
//指定的scope上实例化bean
scopeName = mbd.getScope();
Scope var21 = (Scope)this.scopes.get(scopeName);
if(var21 == null) {
throw new IllegalStateException("No Scope registered for scope \'" + scopeName + "\'");
}
try {
Object var22 = var21.get(beanName, new ObjectFactory() {
public Object getObject() throws BeansException {
AbstractBeanFactory.this.beforePrototypeCreation(beanName);
Object var2;
try {
var2 = AbstractBeanFactory.this.createBean(beanName, mbd, args);
} finally {
AbstractBeanFactory.this.afterPrototypeCreation(beanName);
}
return var2;
}
});
bean = this.getObjectForBeanInstance(var22, name, beanName, mbd);
} catch (IllegalStateException var18) {
throw new BeanCreationException(beanName, "Scope \'" + scopeName + "\' is not active for the current thread; " + "consider defining a scoped proxy for this bean if you intend to refer to it from a singleton", var18);
}
}
}
//检查需要的类型是否符合bean的实际类型
if(requiredType != null && bean != null && !requiredType.isAssignableFrom(bean.getClass())) {
throw new BeanNotOfRequiredTypeException(name, requiredType, bean.getClass());
} else {
return bean;
}
}
对于Spring加载bean的过程,大致分为以下几步:
转换对应beanName
这里传入的参数name不一定就是beanName,有可能是别名或FactoryBean,所以需要进行一系列的解析,这些解析内容包括如下内容
去除FactoryBean的修饰符,也就是如果name=”&aa”,那么会首先去除&而使name=”aa”
取指定alias所表示的最终beanName,例如别名A指向名称为B的bean则返回B;若别名A指向别名B,别名B又指向名称为C的bean则返回C
尝试从缓存中加载单例
单例在Spring的同一个容器内只会被创建一次,后续再获取bean,就直接从单例缓存中获取了。这里只是尝试加载,首先尝试从缓存中加载,如果加载不成功则再次尝试从singletonFactories中加载,因为在创建单例bean的时候会存在依赖注入的情况,而在创建依赖的时候为了避免循环依赖,在Spring中创建bean的原则是不等bean创建完成就会将创建bean的ObjectFactory提早曝光加入到缓存中,一旦下一个bean创建时候需要依赖上一个bean则直接使用ObjectFactory
bean的实例化
如果从缓存中得到了bean的原始状态,则需要对bean进行实例化,这里有必要强调一下,在缓存中记录的只是最原始的bean状态,并不一定是我们最终想要的bean
原型模式的依赖检查
只有在单例情况下才会尝试解决循环依赖,如果存在A中有B的属性,B中有A的属性,那么当依赖注入的时候,就会产生当A还未创建完的时候因为对于B的创建再次返回创建A,造成循环依赖,也就是情况:isPrototypeCurrentlyInCreation(beanName)判断true
检测parentBeanFactory
从代码上来看,如果缓存没有数据的话直接转到父类工厂上去加载,!this.containsBeanDefinition(beanName检测如果当前加载的XML配置文件中不包含beanName所对应的配置,就只能到parentBeanFactory去尝试,然后再去递归的调用getBean方法
将存储XML配置文件的GernericBeanDefinition转换为RootBeanDefinition
因为从XML配置文件中读取到的Bean信息是存储在GernericBeanDefinition中的,但是所有的Bean后续处理都是针对于RootBeanDefinition的,所以这里需要进行一个转换,转换的同时如果父类bean不为空的话,则会一并合并父类属性
寻找依赖
因为bean的初始化过程很可能会用到某些属性,而某些属性很可能是动态配置的,并且配置成依赖于其他的bean,那么这个时候就有必要先加载依赖的bean,所以,在Spring的加载顺寻中,在初始化某一个bean的时候首先会初始化这个bean所对应的依赖
针对不同的scope进行bean的创建
在Spring中存在着不同的scope,其中默认的是singleton,但是还有些其他的配置诸如prototype、request之类的,在这个步骤中,Spring会根据不同的配置进行不同的初始化策略
类型转换
程序到这里返回bean后已经基本结束了,通常对该方法的调用参数requiredType是为空的,但是可能会存在这样的情况,返回的bean其实是个Spring,但是requiredType却传入Integer类型,那么这时候本步骤就会起作用了,它的功能是将返回的bean转换为requiredType所指定的类型,当然,Spring转换为Integer是最简单的一种转换,在Spring中提供了各种各样的转换器,用户也可以自己扩展转换器来满足需求
一般情况下, Spring通过反射机制利用bean的class属性指定实现类来实例化bean,在某些情况下,实例化bean过程比较复杂,如果按照传统的方式,则需要在
Spring为此提供了一个org.Springframework.bean.factory.FactoryBean的工厂类接口,用户可以通过实现该接口定制实例化bean的逻辑
FactoryBean接口对于Spring框架来说占有重要的地位,Spring自身就提供了70多个FactoryBean的实现,它们隐藏了实例化一些复杂bean的细节,给上层应用带来了便利,从Spring 3.0开始,FactoryBean开始支持泛型,即接口声明改为FactoryBean
package org.springframework.beans.factory;
public interface FactoryBean {
T getObject() throws Exception;
Class> getObjectType();
boolean isSingleton();
}
在该接口中还定义了以下3个方法:
当配置文件中
public class Car{
private int maxSpeed;
private String brand;
private double price;
//get/set方法
}
如果用FactoryBean的方式实现就会灵活一些,下例通过逗号分隔符的方式一次性地为Car的所有属性指定配置值
public class CarFactoryBean implements FactoryBean{
private String carInfo;
public Car getObject() throws Exceptino {
Car car = new Car();
String[] infos = carInfo.split(",");
car.setBrand(infos[0]);
car.setMaxSpeed(Integer.valueOf(infos[1]));
car.setPrice(Double.valueOf(infos[2]));
return car;
}
public class getObjectType(){
return Car.class;
}
public boolean isSingleton(){
return false;
}
public String getCarInfo(){
return this.carInfo;
}
//接受逗号分割符设置属性信息
public void setCarInfo(String carInfo){
this.carInfo = carInfo;
}
}
有了这个CarFactoryBean后,就可以在配置文件中使用下面这种自定义的配置方式配置CarBean了
当调用getBean(“car”)时,Spring通过反射机制发现CarFactoryBean实现了FactoryBean的接口, 这时Spring容器就调用接口方法CarFactoryBean#getObject()方法返回,如果希望获取CarFactoryBean的实例,则需要在使用getBean(beanName)方法时在beanName前显示的加上”&”前缀,例如getBean(“&car”)
public Object getSingleton(String beanName) {
//参数true设置标识允许早期依赖
return this.getSingleton(beanName, true);
}
protected Object getSingleton(String beanName, boolean allowEarlyReference) {
//检查缓存中是否存在实例
Object singletonObject = this.singletonObjects.get(beanName);
if(singletonObject == null) {
//如果为空,则锁定全局变量并进行处理
Map var4 = this.singletonObjects;
synchronized(this.singletonObjects) {
//如果此bean正在加载则不处理
singletonObject = this.earlySingletonObjects.get(beanName);
if(singletonObject == null && allowEarlyReference) {
//当某些方法需要提前初始化的时候则会调用addSingletonFactory方法将对应的ObjectFactory初始化策略存储在singletonFactories
ObjectFactory singletonFactory = (ObjectFactory)this.singletonFactories.get(beanName);
if(singletonFactory != null) {
//调用预先设定的getObject方法
singletonObject = singletonFactory.getObject();
//记录在缓存中,earlySingletonObjects和singletonFactories互斥
this.earlySingletonObjects.put(beanName, singletonObject);
this.singletonFactories.remove(beanName);
}
}
}
}
return singletonObject != NULL_OBJECT?singletonObject:null;
}
这个方法首先尝试从singletonObjects里面获取实例,如果获取不到再从earlySingletonObjects里面获取,如果还获取不到,再尝试从singletonFactories里面获取beanName对应的ObjectFactory,然后调用这个ObjectFactory的getObject来创建bean,并放到earlySingletonObjects里面去,并且从singletonFactories里面remove掉这个ObjectFactory,而对于后续的所有内存操作都只为了循环依赖检测时候使用,也就是在allowEarlyReference为true的情况下才会使用
这里涉及用于存在bean的不同map,说明如下:
在getBean方法中,getObjectForBeanInstance是个高频率使用的方法,无论是从缓存中获得bean还是根据不同的scope策略加载bean。总之,得到bean的实例后要做的第一步就是调用这个方法来检测一下正确性,其实就是用于检测当前bean是否是FactoryBean类型的bean,如果是,那么需要调用该bean对应的FactoryBean实例中的getObject()作为返回值
无论是从缓存中获取到的bean还是通过不同的scope策略加载的bean都只是最原始的bean状态,并不一定是最终想要的bean,如需要对工厂bean进行处理,那么这里得到的其实是工厂bean的初始状态, 但是我们真正需要的是工厂bean中定义的factory-method方法中返回的bean,而getObjectForBeanInstance方法就是完成这个工作的
protected Object getObjectForBeanInstance(Object beanInstance, String name, String beanName, RootBeanDefinition mbd) {
//如果指定的name是工厂相关(以&为前缀)且beanInstance又不是FactoryBean类型则验证不通过
if(BeanFactoryUtils.isFactoryDereference(name) && !(beanInstance instanceof FactoryBean)) {
throw new BeanIsNotAFactoryException(this.transformedBeanName(name), beanInstance.getClass());
} else if(beanInstance instanceof FactoryBean && !BeanFactoryUtils.isFactoryDereference(name)) {
//加载FactoryBean
Object object = null;
if(mbd == null) {
//尝试从缓存中加载bean
object = this.getCachedObjectForFactoryBean(beanName);
}
//到这里已经明确知道beanInstance一定是FactoryBean类型
if(object == null) {
FactoryBean factory = (FactoryBean)beanInstance;
//containsBeanDefinition检测BeanDefinitionMap中也就是在所有已经加载的类中检测是否定义beanName
if(mbd == null && this.containsBeanDefinition(beanName)) {
//将存储XML配置文件的GernericBeanDefinition转换为RootBeanDefinition,如果指定BeanName是子Bean的话同时会合并父类的相关属性
mbd = this.getMergedLocalBeanDefinition(beanName);
}
//是否是用户定义的而不是应用程序本身定义的
boolean synthetic = mbd != null && mbd.isSynthetic();
object = this.getObjectFromFactoryBean(factory, beanName, !synthetic);
}
return object;
} else {
return beanInstance;
}
}
getObjectForBeanInstance中所做的工作如下:
将从Factory中解析bean的工作委托给getObjectFromFactoryBean
protected Object getObjectFromFactoryBean(FactoryBean factory, String beanName, boolean shouldPostProcess) {
if(factory.isSingleton() && this.containsSingleton(beanName)) {
//如果是单例模式
synchronized(this.getSingletonMutex()) {
Object object = this.factoryBeanObjectCache.get(beanName);
if(object == null) {
object = this.doGetObjectFromFactoryBean(factory, beanName, shouldPostProcess);
this.factoryBeanObjectCache.put(beanName, object != null?object:NULL_OBJECT);
}
return object != NULL_OBJECT?object:null;
}
} else {
return this.doGetObjectFromFactoryBean(factory, beanName, shouldPostProcess);
}
}
这个方法中只做了一件事情,就是返回的bean如果是单例的,那就必须保证全局唯一,同时,也因为是单例的,所以不必重复创建,可以使用缓存来提高性能,也就是说已经加载过就要记录下来以便于下次复用, 否则的话就直接获取了
doGetObjectFromFactoryBean方法的代码如下:
private Object doGetObjectFromFactoryBean(final FactoryBean factory, String beanName, boolean shouldPostProcess) throws BeanCreationException {
Object object;
try {
//需要权限验证
if(System.getSecurityManager() != null) {
AccessControlContext ex = this.getAccessControlContext();
try {
object = AccessController.doPrivileged(new PrivilegedExceptionAction() {
public Object run() throws Exception {
return factory.getObject();
}
}, ex);
} catch (PrivilegedActionException var8) {
throw var8.getException();
}
} else {
//直接调用getObject方法
object = factory.getObject();
}
} catch (FactoryBeanNotInitializedException var9) {
throw new BeanCurrentlyInCreationException(beanName, var9.toString());
} catch (Throwable var10) {
throw new BeanCreationException(beanName, "FactoryBean threw exception on object creation", var10);
}
if(object == null && this.isSingletonCurrentlyInCreation(beanName)) {
throw new BeanCurrentlyInCreationException(beanName, "FactoryBean which is currently in creation returned null from getObject");
} else {
if(object != null && shouldPostProcess) {
try {
//调用ObjectFactory的后处理器
object = this.postProcessObjectFromFactoryBean(object, beanName);
} catch (Throwable var7) {
throw new BeanCreationException(beanName, "Post-processing of the FactoryBean\'s object failed", var7);
}
}
return object;
}
}
doGetObjectFromFactoryBean实现了从FactoryBean中对应的getObject方法得到bean,但是得到后并没有立即返回,而是做了些后处理的操作
protected Object postProcessObjectFromFactoryBean(Object object, String beanName) {
return this.applyBeanPostProcessorsAfterInitialization(object, beanName);
}
public Object applyBeanPostProcessorsAfterInitialization(Object existingBean, String beanName) throws BeansException {
Object result = existingBean;
Iterator var5 = this.getBeanPostProcessors().iterator();
while(var5.hasNext()) {
BeanPostProcessor beanProcessor = (BeanPostProcessor)var5.next();
result = beanProcessor.postProcessAfterInitialization(result, beanName);
if(result == null) {
return result;
}
}
return result;
}
在Spring获取bean的规则中有这样一条:尽可能保证所有bean初始化后都会调用注册的BeanPostProcessor的postProcessAfterInitialization方法进行处理,在实际开发过程中大可以针对此特性设计自己的业务逻辑
如果缓存中不存在已经加载的单例bean就需要从头开始bean的加载过程了,而Spring中使用getSingleton的重载方法实现bean的加载过程
public Object getSingleton(String beanName, ObjectFactory singletonFactory) {
Assert.notNull(beanName, "\'beanName\' must not be null");
Map var3 = this.singletonObjects;
//全局变量需要同步
synchronized(this.singletonObjects) {
//首先检查对应的bean是否已经加载过,因为singleton模式其实就是复用以前创建的bean,所以这一步是必须的
Object singletonObject = this.singletonObjects.get(beanName);
//如果为null才可以进行singleton的bean的初始化
if(singletonObject == null) {
if(this.singletonsCurrentlyInDestruction) {
throw new BeanCreationNotAllowedException(beanName, "Singleton bean creation not allowed while the singletons of this factory are in destruction (Do not request a bean from a BeanFactory in a destroy method implementation!)");
}
if(this.logger.isDebugEnabled()) {
this.logger.debug("Creating shared instance of singleton bean \'" + beanName + "\'");
}
this.beforeSingletonCreation(beanName);
boolean recordSuppressedExceptions = this.suppressedExceptions == null;
if(recordSuppressedExceptions) {
this.suppressedExceptions = new LinkedHashSet();
}
try {
//初始化bean
singletonObject = singletonFactory.getObject();
} catch (BeanCreationException var13) {
BeanCreationException ex = var13;
if(recordSuppressedExceptions) {
Iterator var8 = this.suppressedExceptions.iterator();
while(var8.hasNext()) {
Exception suppressedException = (Exception)var8.next();
ex.addRelatedCause(suppressedException);
}
}
throw ex;
} finally {
if(recordSuppressedExceptions) {
this.suppressedExceptions = null;
}
this.afterSingletonCreation(beanName);
}
//加入缓存
this.addSingleton(beanName, singletonObject);
}
return singletonObject != NULL_OBJECT?singletonObject:null;
}
}
上述代码中其实是使用了回调方法,使得程序可以在单例创建的前后做一些准备及处理操作,而真正获取单例bean的方法其实并不是再此方法中实现的,其实现逻辑是在ObjectFactory类型的实例singletonFactory中实现的,而这些准备及处理操作包括如下内容
加载单例前记录加载状态
beforeSingletonCreation方法是个空实现,其目的是:记录加载状态,也就是通过this.singletonsCurrentlyInCreation.add(beanName)将当前正在创建的bean记录在缓存中,这样便可以对循环依赖进行检测
protected void beforeSingletonCreation(String beanName) {
if(!this.singletonsCurrentlyInCreation.add(beanName)) {
throw new BeanCurrentlyInCreationException(beanName);
}
}
通过调用参数传入的ObjectFactory的个体Object方法实例化bean
加载单例后的处理方法调用
当bean加载结束后需要移除缓存中对该bean的正在加载状态的记录
protected void afterSingletonCreation(String beanName) {
if(!this.singletonsCurrentlyInCreation.remove(beanName)) {
throw new IllegalStateException("Singleton \'" + beanName + "\' isn\'t currently in creation");
}
}
将结果记录至缓存并删除加载bean过程中所记录的各种辅助状态
protected void addSingleton(String beanName, Object singletonObject) {
Map var3 = this.singletonObjects;
synchronized(this.singletonObjects) {
this.singletonObjects.put(beanName, singletonObject != null?singletonObject:NULL_OBJECT);
this.singletonFactories.remove(beanName);
this.earlySingletonObjects.remove(beanName);
this.registeredSingletons.add(beanName);
}
}
返回处理结果
bean的加载逻辑其实是在传入的ObjectFactory类型的参数singletonFactory中定义的,反推参数的获取,得到如下代码:
sharedInstance = getSingleton(beanName, new ObjectFactory
ObjectFactory的核心部分其实只是调用了createBean的方法