循环依赖其实就是循环引用,也就是两个或则两个以上的 Bean 互相持有对方,最终造成闭环。好比A依赖于B,B依赖于C,C又依赖于A。以下图:java
若是在平常开发中咱们用new 对象的方式发生这种循环依赖的话程序会在运行时一直循环调用,直至内存溢出报错。下面说一下Spring是若是解决循环依赖的。spring
注意,这里不是函数的循环调用,是对象的相互依赖关系。循环调用其实就是一个死循环,除非有终结条件。缓存
根据 spring 中 Bean 的注入方式:构造器注入方式,属性注入方式(单例和多例),咱们分别来说解循环依赖的处理方式。app
Spring容器会将每个正在建立的 Bean 标识符放在一个“当前建立Bean池”中,Bean标识符在建立过程当中将一直保持在这个池中,所以若是在建立Bean过程当中发现本身已经在“当前建立Bean池”里时将抛出BeanCurrentlyInCreationException
异常表示循环依赖;而对于建立完毕的Bean将从“当前建立Bean池”中清除掉。函数
首先咱们先初始化三个Bean。学习
StudentA 测试
public class StudentA { private StudentB studentB ; public void setStudentB(StudentB studentB) { this.studentB = studentB; } public StudentA() { } public StudentA(StudentB studentB) { this.studentB = studentB; } }
StudentB ui
public class StudentB { private StudentC studentC ; public void setStudentC(StudentC studentC) { this.studentC = studentC; } public StudentB() { } public StudentB(StudentC studentC) { this.studentC = studentC; } }
StudentC this
public class StudentC { private StudentA studentA ; public void setStudentA(StudentA studentA) { this.studentA = studentA; } public StudentC() { } public StudentC(StudentA studentA) { this.studentA = studentA; } }
OK,上面是很基本的3个类,StudentA 有参构造是 StudentB 。StudentB 的有参构造是 StudentC,StudentC 的有参构造是 StudentA ,这样就产生了一个循环依赖的状况,咱们都把这三个 Bean 交给 Spring 管理,并用有参构造实例化。.net
<bean id="a" class="com.zfx.student.StudentA"> <constructor-arg index="0" ref="b"></constructor-arg> </bean> <bean id="b" class="com.zfx.student.StudentB"> <constructor-arg index="0" ref="c"></constructor-arg> </bean> <bean id="c" class="com.zfx.student.StudentC"> <constructor-arg index="0" ref="a"></constructor-arg> </bean>
下面是测试类:
public class Test { public static void main(String[] args) { ApplicationContext context = new ClassPathXmlApplicationContext("com/zfx/student/applicationContext.xml"); //System.out.println(context.getBean("a", StudentA.class)); } }
执行结果报错信息为:
Caused by: org.springframework.beans.factory.BeanCurrentlyInCreationException: Error creating bean with name 'a': Requested bean is currently in creation: Is there an unresolvable circular reference?
若是你们理解开头那句话的话,这个报错应该不惊讶,Spring容器先建立单例StudentA,StudentA依赖StudentB,而后将A放在“当前建立Bean池”中,此时建立 StudentB,StudentB 依赖 StudentC ,而后将B放在“当前建立Bean池”中,此时建立StudentC,StudentC又依赖StudentA, 可是,此时Student已经在池中,因此会报错,由于在池中的Bean都是未初始化完的,因此会依赖错误 (初始化完的Bean会从池中移除)。
若是要说setter方式注入的话,咱们最好先看一张Spring中Bean实例化的图
如图中前两步骤得知:Spring是先将Bean对象实例化以后再设置对象属性的
修改配置文件为set方式注入
<!--scope="singleton"(默认就是单例方式) --> <bean id="a" class="com.zfx.student.StudentA" scope="singleton"> <property name="studentB" ref="b"></property> </bean> <bean id="b" class="com.zfx.student.StudentB" scope="singleton"> <property name="studentC" ref="c"></property> </bean> <bean id="c" class="com.zfx.student.StudentC" scope="singleton"> <property name="studentA" ref="a"></property> </bean>
下面是测试类:
public class Test { public static void main(String[] args) { ApplicationContext context = new ClassPathXmlApplicationContext("com/zfx/student/applicationContext.xml"); System.out.println(context.getBean("a", StudentA.class)); } }
打印结果为:
com.zfx.student.StudentA@1fbfd6
为何用set方式就不报错了呢 ?
咱们结合上面那张图看,Spring先是用构造实例化Bean对象 ,此时Spring会将这个实例化结束的对象放到一个Map中,而且Spring提供了获取这个未设置属性的实例化对象引用的方法。 结合咱们的实例来看,当Spring实例化了StudentA、StudentB、StudentC后,紧接着会去设置对象的属性,此时StudentA依赖StudentB,就会去Map中取出存在里面的单例StudentB对象,以此类推,不会出来循环的问题喽。
下面是Spring源码中的实现方法。如下的源码在Spring的Bean包中的DefaultSingletonBeanRegistry.java
类中:
/** Cache of singleton objects: bean name --> bean instance(缓存单例实例化对象的Map集合) */ private final Map<String, Object> singletonObjects = new ConcurrentHashMap<String, Object>(64); /** Cache of singleton factories: bean name --> ObjectFactory(单例的工厂Bean缓存集合) */ private final Map<String, ObjectFactory> singletonFactories = new HashMap<String, ObjectFactory>(16); /** Cache of early singleton objects: bean name --> bean instance(早期的单身对象缓存集合) */ private final Map<String, Object> earlySingletonObjects = new HashMap<String, Object>(16); /** Set of registered singletons, containing the bean names in registration order(单例的实例化对象名称集合) */ private final Set<String> registeredSingletons = new LinkedHashSet<String>(64); /** * 添加单例实例 * 解决循环引用的问题 * Add the given singleton factory for building the specified singleton * if necessary. * <p>To be called for eager registration of singletons, e.g. to be able to * resolve circular references. * @param beanName the name of the bean * @param singletonFactory the factory for the singleton object */ protected void addSingletonFactory(String beanName, ObjectFactory singletonFactory) { Assert.notNull(singletonFactory, "Singleton factory must not be null"); synchronized (this.singletonObjects) { if (!this.singletonObjects.containsKey(beanName)) { this.singletonFactories.put(beanName, singletonFactory); this.earlySingletonObjects.remove(beanName); this.registeredSingletons.add(beanName); } } }
修改配置文件为:
<bean id="a" class="com.zfx.student.StudentA" scope="prototype"> <property name="studentB" ref="b"></property> </bean> <bean id="b" class="com.zfx.student.StudentB" scope="prototype"> <property name="studentC" ref="c"></property> </bean> <bean id="c" class="com.zfx.student.StudentC" scope="prototype"> <property name="studentA" ref="a"></property> </bean>
scope="prototype" 意思是 每次请求都会建立一个实例对象。二者的区别是:有状态的bean都使用 prototype 做用域,无状态的通常都使用singleton单例做用域。
测试用例:
public class Test { public static void main(String[] args) { ApplicationContext context = new ClassPathXmlApplicationContext("com/zfx/student/applicationContext.xml"); //此时必需要获取Spring管理的实例,由于如今scope="prototype" 只有请求获取的时候才会实例化对象 System.out.println(context.getBean("a", StudentA.class)); } }
打印结果:
Caused by: org.springframework.beans.factory.BeanCurrentlyInCreationException: Error creating bean with name 'a': Requested bean is currently in creation: Is there an unresolvable circular reference?
为何原型模式就报错了呢 ?
对于“prototype”做用域Bean,Spring容器没法完成依赖注入,由于“prototype”做用域的Bean,Spring容器不进行缓存,所以没法提早暴露一个建立中的Bean。
来源于:学习园